In thyroid diseases, it is known that the activity of serum angiotensin-converting enzyme(ACE) increases in Graves’ hyperthyroidism,^1–4) and that it becomes normal in euthyroid patients after treatment. Even though the mechanism of increased serum ACE activity is still unknown, Yotsumoto et al.³⁾ and Nakamura et al.²⁾ reported that the amount of thyroid hormone and the ACE activity in serum had changed proportionately. And other in vitro reports^5–7) suggested that the thyroid hormone has an influence upon the secretion of ACE from tissues to blood and upon the activity of ACE in the cultured cells in vitro or in an animal. Therefore, it is now thought that the increase of serum ACE activity in the thyroid disease is generally caused by the increase of the thyroid hormone per se^2–3) rather than by the thyrotropin receptor related antibody, the cause of hyperthyroidism. If serum ACE activity is related to the functional status of the thyroid, serum ACE activity in the patients with hypothyroidism would have been found to be low, but according to Nakamura et al.²⁾ and Silverstein et al.⁴⁾ there was no difference between the activities of hypothyroid patients and those of the normal control group. Mean while, as the autoantibody toward TSH receptor of his own has been regarded as an autoimmune cause of the elevation of the thyroid function, it is necessary to examine the relationship of elevated serum ACE activity in Graves’ disease to thyrotropin receptor related antibody, although there have been no reports about these possibilities. We have examined the changes of thyroid functions in the patients with Graves’ disease in hyperthyroidism and traced the serum ACE activity and thyrotropin-binding-inhibitory immunoglobulin (TBII) activity and have compared those with each other.

Serum ACE activity according to age and sex in normal subjects has been reported variably by several authorities. According to Lieberman,⁸⁾ the serum ACE activity was higher in the age group under 20 years old, but Ashutosh et al.⁹⁾ stated the contrary. With regard to sexual differences, Lieberman, Beutler et al.¹⁰⁾ and Nakamura et al.³⁾ said that the activity was higher in men than in women. But Matsuki et al.¹¹⁾ and Hurst et al.¹²⁾ reported that there was no difference in the serum ACE activity between men and women. Our results were similar to these reports in that no differences were detected between older age groups and the young, and between male and female subjects.

Since it has been reported that the serum ACE activities increase in various diseases, it is well known that increased serum ACE activity is not a unique feature of a certain disease. Recently there were a few reports where by the ACE activity increased in proportion to the increase of thyroid function.^2,3,13) Yotsumoto²⁾ et. al. reported that normal serum ACE activity was 30 ± 9 nmoles/min/ml on an average and that the activity in patients with hyperthyroidism was 65 ± 18 nmoles/min/ml. He also said that 80% of the patients with hyperthyroidism in his series showed the increase of ACE activity more than twice the standard deviation range of the normal activity. Nakamura³⁾ et al. reported that the average was 30 nmoles/min/ml in the normal male and 26 nmoles/min/ml in the normal female subjects and 52 nmoles/min/ml in patients with hyperthyroidism. He also pointed out that serum ACE activity in hyperthyroidism was significantly higher than in the normal subjects, although in hypothyroiidism the activity did not show a statistically significant decrease. Our results showed that the serum ACE activity was 23.7 ± 6.0 nmoles/min/ml in the normal control subjects and was 34.5 ± 13.3 nmoles/min/ml in untreated patients with hyperthyroid Graves’ disease. In addition, only 60% of the patients showed the increase of more than one standard deviation over the mean value of normal controls. These differences in the ranges of serum ACE activities from several reports are now thought to be caused mainly by the difference of buffers chosen in the assay and partly by other factors. According to Cushman et al.¹⁴⁾ the serum ACE was reported to show its maximum activity at pH 8.3. But the buffer capacity of the phosphate used is not perfect at pH 8.3. Besides, the phosphate itself inhibits the ACE activity partly, that is, the relative activity was reported to be 0.82 when the phosphate buffer was used as compared with that measured when the borate buffer was used.¹²⁾ In addition, Lieberman⁸⁾ pointed out that when the absorption of hippuric acid produced was more than 0.6, the linearity of absorbance curve began to decline a little and recommended the assay should be repeated with a diluted sample to get the exact result. But according to our preliminary observation (data, not shown), the linearity of the ratio of absorption to concentration was maintained up to 1.5 of absorption. Our result regarding the serial changes of activity are those obtained within the same assay to avoid the possible large range of error of comparison with interassay variance. This would have contributed to the less sensitivity of the measured activity in our series.

In the 15 patients with hypothyroidism, the serum ACE activity showed a significant decrease than the normal control group. This result was different from those of Nakamura et al.³⁾ and Silverstein et al.⁴⁾ These differences are considered to be caused mainly by the difference of the patients who were included, the assay method and the accuracy of measurement. Because the serum lipid contents increase in the patients with hypothyroidism, the possibility that the interference by the lipid contained in blood had an influence upon the measured activity should also be considered.¹⁵⁾ But above all these limitations and the results of our less sensitive assay showed significant decrease in serum ACE activity in the patients with hypothyroidism, it can be infered that the serum ACE activities decrease in a hypothyroid state in proportion. But before extrapolating the results of this report, we should clarify the interference of the serum lipid especially when the blood sample was drawn from patients for the assay after meals. Nakamura et al.³⁾ and Kwon et al.¹³⁾ reported that there was significant correlation between thyroid hormone levels and the serum ACE activity. Nakamura et al.³⁾ included patients with Graves’ disease before treatment and during treatment, and Kwon et al.¹³⁾ excluded the hypothyroid patients with variable ACE activities for the analysis of correlation. We included all the activities measured in concert with thyroid function tests and found a significant correlation between them with the correlation coefficient of about 0.6. Though we intended to ascertain whether the thyrotropin receptor related antibody itself increased the serum ACE activity in the thyroid or the periphery, there was no correlation between serum TBII activities and ACE activities. Because the ACE is not distributed in especially large quantities in the thyroid tissue,^16,17) we could think that the enlarged thyroid did not produce and secrete ACE into the serum according to TBII activities. Though we should take the limitations of TBII in representing abnormal stimulators in Graves’ disease into consideration and the least possibility that TBII would have an effect on the peripheral tissues, it might be suggested that the TBII would have no effect on the changes of the serum ACE activity in Graves’ hyperthyroidism.

Yotsumoto et al.²⁾ observed that in the patients with hyperthyroidism, after measuring both the change of the thyroid hormones and the change of the serum ACE activities during treatment, the serum ACE activities decreased together with the decrease of the thyroid hormones. He also reported that there was a trend toward increase in serum ACE activity about two weeks later when the level of thyroid hormone in blood increased again due to the insufficiency of treatment. In this study, we can observe similar features in the cases of which we can follow up both the thyroid functions and the serum ACE activity. In cases who were given antithyroid drugs alone, in accordance with the normalization of the thyroid function, the serum ACE activity decreased as the thyroid function normalized and it showed the same results in cases of combined treatment with radioiodine. In a case, despite the use of high dose of antithyroid drugs the patient was in hyperfunction for four months and after the additional administration of radioiodine in accordance with the normalization of the thyroid function, the serum ACE activities decreased (Fig. 7). On the other hand, in a case of treatment by radioiodine alone, though the serum TBII activity showed transient fluctuation, the serum ACE activity decreased down to normal range with an improvement of the thyroid functional state (Fig. 6). In addition, the serum ACE activity was yet in the normal range though the function was elevated by the reactivation of the disease in the last month, so there should be a delay in the increase of the serum ACE activity in proportion to the thyroid functional state. We couldn’t find any direct correlation between the change of the serum ACE activity and TBII activity, and not in another case, either (Fig. 8). In brief, there was no direct relationship between the abnormal thyroid stimulators, TBII, and the enzyme activities in patients with Graves’ disease regardless of the regimen. And the serum ACE activity changed a little later than the increase of the thyroid hormone. These facts can be interpreted to be an indirect evidence of an assumption that the major factor in the changes of serum ACE activity is the peripheral thyroid hormone concentration.

Because the hyperthyroidism itself brings about the hemodynamic excess, the assumption that the hemodynamic excess state in hyperthyroidism should increase the ACE activity by changing the peripheral circulation including the pulmonary circulation is possible as suggested in chronic liver disease.¹¹⁾ But, since there has been no report that serum ACE activity increased in the other diseases known to cause severe hemodynamic excess, we could pose less possibility that this would be the main mechanism increasing the serum ACE activity in hyperthyroidism. On the other hand, the evidence that thyroid hormones will do much directly and/or indirectly by inducing the ACE secretion from peripheral tissues to the serum has been sought in the animal experiments and in vitro.

Friedland et al.¹⁹⁾ reported that the particulate ACE of endothelial cells has the same activity and characteristics as the serum ACE when the soluble ACE was separated by treating with the proteolytic enzyme of trypsin the particulate ACE of vascular endothelial cells of rabbit. From the fact that the membrane bound ACE and the serum ACE are similar in their physicochemical characteristics and from another observation by immunofluorescence that ACE is bound to the membrane of the endothelial cell of the peripheral blood vessels,^20,23) it can be infered that the membrane bound ACE in the endothelial cells is secreted into serum and shall make the serum ACE activity in certain conditions. Silverstein et al. concluded that the thyroid hormone could not induce the new production of ACE showing that the triiodothyronine was not able to induce ACE in the supernatant of the cultured alveolar macrophage cells,⁴⁾ though the enzyme could be induced in the supernatant fluid by the corticosteroid.⁵⁾ These experimental reports cited and our clinical results suggest that, even if the thyroid hormone does not induce the production of ACE in man from peripheral endothelial cells, it will play an important role directly and/or indirectly in the increase or decrease of the serum ACE activities inducing the increased or decreased secretion of ACE into the serum. In conclusion, our results that the serum ACE activities change in proportion to the thyroid hormone levels, from the hyperthyroid state to the normal or even hypothyroid state, was regarded as evidence of an assumption that thyroid hormone will play a role in the secretion and consequent increase in the activity of ACE in the serum.