| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Departments of Internal Medicine I (Endocrinology and Metabolism) (P.H., E.B., R.Z.) and Gynecology and Obstetrics (J.W.), University of Heidelberg, D-69115 Heidelberg, Germany
Address all correspondence and requests for reprints to: Dr. Peter Heilmann, Department of Internal Medicine I (Endocrinology and Metabolism), Luisenstrasse 5, Gebäude 8, D-69115 Heidelberg, Germany. E-mail: Peter_Heilmann{at}med.uni-heidelberg.de
Abstract
In this study we evaluated the activity of renal 11ß-hydroxysteroid dehydrogenase type 2 (11ß-HSD2) in patients with pregnancy-induced hypertension (PIH). A reduction of the activity of 11ß-HSD2 leads to pseudohyperaldosteronism due to insufficient interconversion of cortisol to its inactive 11-oxo-metabolite cortisone in the renal tubulus cell. We measured urinary free cortisol and cortisone in patients with and without PIH and calculated the urinary free cortisol to free cortisone ratio, which is well accepted as a correlate of the activity of renal 11ß-HSD2. One hundred twenty-six pregnant women were included. Fifty-nine patients had PIH (mean age 31.5 ± 4.4 yr, blood pressure 158.7 ± 16.0/100.8 ± 9.5 mm Hg), and 67 were normotensive (mean age 29.4 ± 4.6, blood pressure 112.6 ± 8.9/68.8 ± 8.6 mm Hg). The excretion rate of cortisol was increased in the PIH group (138.8 ± 93.0 vs. 106.5 ± 65.4 nmol/d, P = 0.027), whereas excretion rate of cortisone was similar (362.9 ± 254.1 vs. 366.5 ± 221.7 nmol/d, P = 0.933). The free cortisol to free cortisone ratio was significantly higher in the PIH group (0.47 ± 0.25 vs. 0.31 ± 0.12, P < 0.00002). Within this group, the patients with blood pressure in the uppermost quartile had a significantly higher free cortisol to free cortisone ratio than those in the lowest quartile [0.61 ± 0.31 vs. 0.38 ± 0.15 (P = 0.019) for diastolic, 0.60 ± 0.29 vs. 0.35 ± 0.13 (P = 0.012) for systolic, and 0.62 ± 0.32 vs. 0.39 ± 0.16 (P = 0.023) for mean blood pressure, respectively]. We conclude that a reduction of the activity of the 11ß-HSD2 is a relevant factor for the development of PIH. Whether the ratio of urinary free cortisol to free cortisone is a useful risk factor for the development of PIH must be investigated in further prospective studies.
DEVELOPMENT OF HYPERTENSION in pregnancy is still not fully understood. Normally, blood pressure falls soon after conception due to a decrease of peripheral resistance. Therefore, the renin-angiotensin-aldosterone (Aldo) system is activated causing volume expansion and an increase of cardiac output. In preeclampsia, the decrease of peripheral resistance is blunted, possibly due to an alteration of PG synthesis or the sympathetic system, but additional factors are to be considered (1, 2).
Page et al. (3, 4, 5) studied the relevance of novel secreted peptides from the placenta and discovered placental neurokinin B (NKB) (3, 4, 5), which is grossly elevated in pregnancy-induced hypertension (PIH) (4). This peptide plays an important role for uterine contractility and placental perfusion and is involved in vascular changes causing a shunt of maternal blood from certain organs to the uterus and placenta (5). Therefore, it is likely that the regulation of the secretion of NKB and other neurokinins is involved in the development of PIH and other symptoms of preeclampsia.
Other possible factors for the development of PIH may be due to the involvement of the mineralocorticoid axis. Geller et al. (6) described a mutation of the MR, S810L, which is associated with severe hypertension. Normally, progesterone has an antagonistic activity at the wild-type MR. In contrast, the mutant-type MR shows an intrinsic activation by progesterone and other 21-deoxysteroids. Moreover, classic antagonists act as potent agonists, e.g. spironolactone. Geller reports that two of the carriers of the MRL810 mutation have undergone five pregnancies. All these pregnancies were complicated by severe hypertension, probably due to an activation of the mutant MRL810 receptor by progesterone, which is markedly increased during pregnancy (6).
In vitro, cortisol (F) and Aldo have the same affinity to the MR (7, 8). In vivo, interconversion of F and its inactive 11-keto-metabolite cortisone (E) takes place, catalyzed by 11 ß-hydroxysteroid dehydrogenases (11 ß-HSDs). Until now, two isozymes were known. Type 1 (11ß-HSD1) is a low affinity reductase activating E to F in many tissues, mainly in the liver; and type 2 (11ß-HSD2), a high affinity oxidase, inactivates F to E in the renal tubulus cell and in the placenta and is therefore responsible for the selective access of Aldo to the MR in vivo (9). A congenital deficiency of 11ß-HSD2 leads to a decreased protection of the MR against F and thus to the syndrome of apparent mineralocorticoid excess (AME) with Na+ retention, hypokalemia, hypertension, and suppression of the renin-angiotensin-Aldo axis as described by New et al. (10), Ulick et al. (11), Edwards et al. (12), Shackleton et al. (13), and others (14, 15, 16, 17). Moreover, Ulick et al. described a second form of the AME syndrome (type 2) (18). In this form, interconversion of F and E is impaired in both directions, and half-life of F is prolonged due to an impairment of ring A reduction, shown by a markedly decreased ring A reduction constant (THF + allo-THF/F). Ulick et al. (19) studied ring A reduction constant in patients with AME type 1 and found that it is prolonged in these patients, too. Therefore, a defect in ring A reduction has been suggested as a major error in both types of AME. Nevertheless, deficiency of the renal 11ß-HSD2 plays a key role for the development of hypertension in AME syndrome because of the insufficient inactivation of F in the kidneys.
The role of 11 ß-HSDs in blood pressure regulating or common diseases such as obesity or polycystic ovary syndrome is under discussion (20, 21, 22, 23, 24). The monitoring of the in vivo activity of 11 ß-HSDs is still a challenge. There is current consensus that the ratio of free F to free E (fF/fE ratio) in urine is a good marker for the activity of 11 ß-HSD2 in vivo (25, 26, 27), but it is still very complex to measure this parameter directly. The aim of our study was to evaluate a possibly relevant reduction of the activity of 11 ß-HSD2 in patients with PIH by measuring this parameter using a method that is suitable under routine conditions.
Experimental subjects and protocols
A total of 126 pregnant women were included. According to a consultation of the local Ethical Committee, the course and the protocols of the study were carefully explained to them before they consented to participate. None of the subjects had obvious evidence of other metabolic, endocrine, or hepatic disease. All patients were normotensive before pregnancy. A total of 59 patients had PIH (PIH group), and 67 were normotensive (controls). Patients were included in the PIH group if increased blood pressure was detected on at least three different days. Increased blood pressure was defined as systolic blood pressure greater than or equal to 140 mm Hg and diastolic blood pressure greater than or equal to 90 mm Hg. Patients were included in the control group if they were normotensive on three consecutive occasions. During follow-up, no patient of the control group had hypertensive blood pressure according to the above-mentioned criteria.
Systolic, diastolic, as well as mean blood pressure (RR) was
significantly higher in the PIH group than in the controls
(P < 0.00001). There were no significant differences
between the two groups concerning age and days of pregnancy (Table 1
). In the PIH group,49 patients
had proteinuria (0.3–1.0 g/liter in 27 patients, 1.0–5.0 g/liter in
12 patients, and >5.0 g/liter in 10 patients).
|
Urine samples
The urine specimens were stored at 4 C during the 24-h period of collection. From these samples, aliquots of about 10 ml were stored frozen (-30 C) until analysis. No more sample preparation steps were necessary before simultaneous determination of urinary free F, E, and their 20-dihydrometabolites.
Solvents
The following solvents were used for cleanup and chromatography: acetonitrile, methanol, deionized water, 20 mmol/liter LiOH, 20 mmol/liter phosphoric acid, 20 mmol/liter citric acid, and 50 mmol/liter trifluoracetic acid. Aqueous solvents were freshly prepared before analysis, and organic solvents were of analytical grade and purchased from Merck & Co. (Darmstadt, Germany). All solvents were degazed by purging with helium.
Procedures
Urinary free F, free E, and their free 20-dihydrometabolites were determined using the fully automated Automatic Liquid Chromatographic Analyzer system, which has been described in detail elsewhere (28, 29). The Automatic Liquid Chromatographic Analyzer system is an analytical technology that is based on liquid-solid phase processes and provides the automatic and selective analysis of substances accessible by chromatographic processes. It consists of the sample cleanup unit, the chemical modulator unit, and the analytical unit. The cleanup unit consists of a pump with a ternary mixing device, an autosampler, and a precolumn (precolumn 1) packed with reversed-phase material (PLRP-I, 15 µm; Hamilton, Reno, NV). The chemical modulator unit consists of the mixing chamber, a second precolumn (precolumn 2) packed with octadecyl-silica (5 µm; Shandon Southern Products, Cheshire, U.K.), and a pump. The analytical unit consists of a second pump with a ternary mixing device, the analytical column packed with octadecyl-silica (5 µm; Shandon), and the UV detector. The whole procedure is time controlled by a microprocessor unit.
The sample is loaded onto precolumn 1 and prepared using different organic gradients and different levels of pH. In the second step, the prepurified F and E-containing fraction is eluted from precolumn 1 and focused on-line on top of precolumn 2 by mixing water into the mixing chamber. The cleaned and focused fraction is then transferred on the analytical column, and F, E, and their 20-dihydrometabolites are separated chromatographically. The methodological details of this fully automated liquid chromatographic technique have been described elsewhere (29). Coefficients of variation range between 1.5 and 5.4% for within-assay precision and between 5.7 and 7.6% for between-assay precision, respectively. Analytical recovery is between 94.3 and 104.2%. Assay sensitivity is 10–15 nmol/liter.
Statistical methods
The results of the excretion rates of the electrolytes, the blood pressure measurements, and the excretion rates of the steroids were compared using the t test. The statistical results have been calculated using the program Excel 97 (Microsoft Corp., Redmond, WA) on an IBM- compatible computer.
Results
Na+/K+ ratio in serum and urine
Serum level of Na+ and K+ were within the normal range in both groups, and no significant differences were detectable. The Na+/K+ ratio in urine was higher in the control group, but this difference was not statistically significant (3.94 ± 2.17 vs. 3.44 ± 1.74, P = 0.16).
Excretion rates of F, E, and 20 
-dihydrocortisol (DHF)
Urinary free F was significantly higher for patients in the PIH
group than for those in the control group (138.8 ± 93.0
vs. 106.5 ± 65.4 nmol/d, P = 0.027),
whereas urinary free E was similar in both groups (362.9 ± 254.1
and 366.5 ± 221.7 nmol/d, respectively, P =
0.933). Urinary free 20 -DHF was higher in the PIH group than in the
controls (410.2 ± 264.5 vs. 314.7 ± 246.9
nmol/d, P = 0.05). Correspondingly, the ratio of DHF/F
was not different in both groups (3.1 ± 1.5 vs.
2.9 ± 1.3, P = 0.465).
fF/fE ratio
The fF/fE ratio was significantly higher in the PIH group than in
the controls (0.47 ± 0.25 vs. 0.31 ± 0.12,
P < 0.00002; Fig. 1).
However, there is an overlap between the fF/fE ratios of the two
groups. The fF/fE ratio was still increased in patients with PIH if
only those patients whose excretion rate of F and E was comparable to
that of the controls were included (a range of the sum of free F and E
(±SD) was calculated for the control group,
P < 0.0005; Fig. 2).
Moreover, in the PIH group, the fF/fE ratio was significantly higher in
those patients with a low excretion of free F plus E than in those
patients with higher excretion of free F plus E (0.64 ± 0.33 in
the lowest quartile vs. 0.34 ± 0.14 in the uppermost
quartile of excretion of F plus E, P < 0.005). In the
control group, there was no such difference (0.30 ± 0.13 in the
lowest quartile vs. 0.27 ± 0.11 in the uppermost
quartile of excretion of F plus E, P = 0.46).
|
|
Correlation to blood pressure
The fF/fE ratio increased with increasing values for diastolic, mean, and systolic blood pressure, but these correlations were not significant (0.05 <P < 0.15).
In patients of the PIH group with a diastolic, mean, or systolic blood
pressure in the uppermost quartile, the fF/fE ratio was higher than in
patients with a blood pressure in the lowest quartile. The difference
was 0.61 ± 0.31 vs. 0.38 ± 0.15
(P = 0.019) for diastolic, 0.60 ± 0.29
vs. 0.35 ± 0.13 (P = 0.012) for
systolic, and 0.62 ± 0.32 vs. 0.39 ± 0.16
(P = 0.023) for mean blood pressure, respectively (Fig. 3).
|
In the last decade, hypertension research has shifted strongly in the direction of molecular genetics. Monogenic hypertensive syndromes have been described that deal with the mineralocorticoid pathway. Two of them depend on an alteration of the mineralocorticoid activity by activation of the MR via ligands other than the classic mineralocorticoid Aldo.
Last year, Geller et al. (6) described a mutation of the MR (S810L), leading to early-onset hypertension with pregnancy-induced exacerbation. This feature is due to a strong activation of the mutant MR by progesterone and other 21-deoxysteroids. Cyclus-dependent hypertensive disorder in women with the mutant MR would be another logical event, but Geller et al. report no data from their patients concerning this possibility.
Another form of monogenic mineralocorticoid hypertension is the so-called syndrome of AME. This syndrome is based on the fact that in vivo selectivity of the MR is not due to the receptor itself, but due to the interconversion of F to E by the 11 ß-HSD2, which therefore protects the MR against F access in mineralocorticoid target tissues such as the renal tubulus cell. An inborn defect of the renal 11 ß-HSD2 reduces this protection and leads to a mineralocorticoid activity of F in vivo (9).
It is under discussion whether mild alteration of the activity of the 11 ß-HSD2 could be responsible for other hypertensive states (9, 20, 24). Walker et al. (30) studied the half-life of F in patients with essential hypertension and described a prolongation in about 20% of these patients. The existence of so-called GALFs (glycyrrhetinic acid-like factors) that are able to inhibit 11 ß -HSD2 has been discussed (24, 31). McCalla et al. (32) found a reduced activity of the placental 11 ß-HSD2 in patients with preeclampsia, an increased level of F in umbilical cord blood, and a significant correlation of the activity of 11 ß-HSD2 and birth weight. Wacker et al. (2) studied patients with PIH and found a situation also comparable to AME, with pseudohyperaldosteronism and suppression of the renin-angiotensin-Aldo system. Walker et al. (31) measured GALFs in patients with PIH and found increased levels, but they did not find a change of the fF/fE-ratio in the urine of these patients.
However, if there is an inhibition or an inborn reduction of the activity of the 11 ß-HSD2 in at least some of the patients with PIH, then this reduction of renal ability to convert F to E should be visible via an increased fF/fE ratio in urine, which is a widely accepted marker for the activity of renal 11 ß-HSD2 (25, 26, 27). Under physiological conditions, the F/E ratio in serum is about 10 and the fF/fE ratio in urine is about 0.2–0.5 (33, 34). This change in the relation between F and E shows the high turnover of F to E in the kidneys.
In our study, the fF/fE ratio in urine was significantly higher in the
PIH group than in the controls. A ratio of more than 0.61 was found
only in patients with PIH. The increased fF/fE ratio might be due to an
overload of 11 ß-HSD2 by an excess of F, as has
been discussed for the situation of ACTH-excess states (33, 35, 36, 37). However, we consider this unlikely for two reasons:
First, in our study the fF/fE ratio was also increased in those
patients whose total excretion of F and E was in the same range as in
the controls (Fig. 2
), and second, in the PIH group the fF/fE ratio was
higher in patients with low total excretion of F plus E, whereas in the
control group there was no such difference. These results within the
PIH group argue against a sole overload by F, although a competitive
inhibition by other metabolites cannot be ruled out. For example,
influences on the activity of the 11 ß-HSD2 by
progesterone and its metabolites, or by GALFs, is under discussion
(31, 38). Progesterone itself seems to be metabolized, at
least in part, by the renal 11 ß-HSD2
(39).
It is difficult to assess whether a mineralocorticoid excess due to an alteration of the activity of the 11 ß-HSD2 plays a role in the development of PIH. The level of sodium and potassium in serum was not significantly different in the two groups. In urine, the ratio of sodium to potassium was lower in the PIH group. This could be a correlate for mineralocorticoid activity, but again, this difference was not significant, possibly due to the fact that no sodium- and potassium-balanced diet was given during collection of urine samples. However, because the aim of the study was to evaluate the possibility of a relevance of the fF/fE ratio in urine as a risk factor for PIH under basal conditions, such a diet was not considered.
The finding of significant differences in the fF/fE ratio within the patients of the PIH group indicate that differences in the activity of 11 ß-HSD2 play a role in the development of PIH. The correlation between blood pressure and fF/fE ratios did not reach a level of significance, probably due to the relatively low number of patients. Nevertheless, the fact that patients with a blood pressure in the uppermost quartile had significantly higher urinary fF/fE ratios in comparison to those patients in the lowest quartile is in good accordance with the theory that an alteration of the activity of 11 ß-HSD2 is of relevance for the level of blood pressure. Alteration of PG synthesis and the sympathetic system have been considered to influence peripheral resistance and blood pressure in pregnancy (1). Page et al. (3, 4, 5) discovered placental NKB and discussed that tachykinins such as NKB play an important role for vascular changes in pregnancy. They proposed that incomplete trophoblast invasion may induce a reflectory increased secretion of NKB leading to further hemodynamic effects and the life-threatening symptoms of preeclampsia. We do not believe that lower activity of 11 ß-HSD2 is the main reason for PIH, but it is thinkable that a decreased inactivation of F in different tissues is an additional factor in the equilibrium of all different effectors that influence peripheral resistance and vascular changes and that this contributes to a shift in blood pressure toward higher levels. Influences of the F/E equilibrium on vasculator tone have been described under experimental conditions (40). In this sense, an alteration of the activity of 11 ß-HSD2 may be an additional risk factor pointing to patients at higher risk for PIH. The fF/fE ratio in urine is a simple and noninvasive method for monitoring the activity of renal 11 ß-HSD2 under routine conditions.
Obviously, urinary fF/fE ratio is not only influenced by the activity of renal 11 ß-HSD2. However, there is a current consensus that urinary fF/fE ratio is the most accurate indicator of both types of AME syndrome. Nevertheless, activity of 11 ß-HSDs in other tissues, especially the activity of placental 11 ß-HSD2, would be interesting with regard to the possibility of local changes of vascular tone. It seems probable that the activity of placental 11 ß-HSD2 is correlated to the activity of the renal enzyme, but this must be elucidated in further studies. However, in our study it was not possible to get placental tissue specimen and to measure the activity of the placental 11 ß-HSD2 in vitro.
From our results, we conclude that the activity of the 11 ß-HSD2 plays a role for the development of PIH. Whether this is due to an inborn reduction of the activity of the 11 ß-HSD2 or whether this reflects an inhibition of this enzyme by other factors remains to be investigated in further studies. The HPLC method we used in our study is a simple method for the simultaneous determination of urinary free F and E. Because it is suitable for the routine laboratory, it could be helpful for further evaluation of this parameter.
Footnotes
Abbreviations: Aldo, Aldosterone; AME, apparent mineralocorticoid excess; DHF, dihydrocortisol; E, cortisone; F, cortisol; fF/fE ration, ratio of free F to free E; GALF, glycyrrhetinic acid-like factor; 11 ß-HSD, 11 ß-hydroxysteroid dehydrogenase; NKB, neurokinin B; PIH, pregnancy-induced hypertension.
Received December 13, 2000.
Accepted July 30, 2001.
References
-hydroxyprogesterone. Endocrinology 137:2308–2314[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  A. E. Michael and A. T. Papageorghiou Potential significance of physiological and pharmacological glucocorticoids in early pregnancy Hum. Reprod. Update, September 1, 2008; 14(5): 497 - 517. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Quinkler and P. M. Stewart Hypertension and the Cortisol-Cortisone Shuttle J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2384 - 2392. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Kajantie, L. Dunkel, U. Turpeinen, U.-H. Stenman, P. J. Wood, M. Nuutila, and S. Andersson Placental 11{beta}-Hydroxysteroid Dehydrogenase-2 and Fetal Cortisol/Cortisone Shuttle in Small Preterm Infants J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 493 - 500. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
