SEARCH

SEARCH BY CITATION

Keywords:

  • acromegaly;
  • growth hormone;
  • kidney;
  • klotho;
  • phosphate

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Abstract.  Sze L, Bernays RL, Zwimpfer C, Wiesli P, Brändle M, Schmid C (Division of Endocrinology and Diabetes, University Hospital, Zurich; Division of Endocrinology and Diabetes, Kantonsspital, Gallen; Department of Neurosurgery, University Hospital, Zurich; and Department of Internal Medicine, Kantonsspital Frauenfeld, Frauenfeld, Switzerland). Excessively high soluble Klotho in patients with acromegaly. J Intern Med 2012; 272: 93–97.

Objectives.  Klotho-deficient mice develop a syndrome resembling accelerated ageing, and genetic variants of Klotho have been associated with human ageing. In humans, serum levels of soluble Klotho decrease with age and with chronic renal failure. The aim of our study was to examine the relationship between excess growth hormone (GH) and serum levels of Klotho in patients with acromegaly, a disease usually caused by a pituitary adenoma, which is associated with high phosphate levels and reduced life expectancy.

Patients and design.  We determined the levels of soluble Klotho, GH and insulin-like growth factor 1 (IGF-1) in serum samples from 24 consecutive patients with acromegaly (nine women/15 men, age 28–76 years) before and after transsphenoidal surgery.

Results.  Soluble Klotho levels were excessively high at baseline (mean ± SEM, 4.2 ± 0.7 ng mL−1) and correlated with GH (r = 0.64), IGF-1 (r = 0.57) and tumour size (r = 0.5). In multiple regression analysis, soluble Klotho was associated with GH after correction for age, gender and levels of creatinine and phosphate (= 0.029). After surgery, GH and IGF-1 levels decreased in all patients (from 26.3 ± 5.2 to 2.6 ± 0.6 μg L−1, P <0.0001, and from 588 ± 35 to 193 ± 12 μg L−1, < 0.001, 0.0001, respectively). Creatinine increased from 71 ± 3 to 80 ± 3 μmol L−1 (< 0.001), and phosphate decreased from 1.37 ± 0.04 to 1.06 ± 0.02 mmol L−1 (< 0.001). The markedly increased preoperative levels of soluble Klotho returned towards normal after surgery (0.7 ± 0.1 ng mL−1, < 0.0001).

Conclusions.  This is the first study to show dramatically increased soluble Klotho levels in an acquired disease in humans. Reversal following tumour removal suggests a causal relation between the GH-producing adenoma and high serum Klotho concentration in acromegaly.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Klotho was identified as a gene that influences lifespan following recognition that disruption of this gene causes a syndrome resembling ageing in mice [1]. Genetic variants of Klotho have been associated with human ageing [2], and a loss of function mutation has been described in a patient with high blood phosphate levels and soft tissue and vascular calcification [3]. Klotho is predominantly expressed in the kidneys and choroid plexus, and expression decreases with chronic renal failure [4]. Klotho protein exists in membrane-bound and secreted soluble forms. Membrane-bound Klotho serves as a coreceptor for fibroblast growth factor 23 (FGF23) [5, 6] and thus has an important role in the renal effects of FGF23 on phosphate transport and calcitriol production. Soluble Klotho (sKlotho) inhibits insulin/insulin-like growth factor 1 (IGF-1) signalling and regulates calcium homoeostasis [7, 8]. A sandwich enzyme-linked immunosorbent assay (ELISA) to measure sKlotho has become available recently [9]. Data are still scarce, but a low concentration of sKlotho has been associated with increasing age, reduced glomerular filtration rate (GFR) and increased risk of mortality and cardiovascular disease [9–11].

Excess growth hormone (GH) production in acromegaly (generally owing to an adenoma of the pituitary) causes pathognomonic clinical signs, distinct metabolic changes and reduced life expectancy. It leads to high levels of serum phosphate in the presence of an increased GFR [12–14] and induces insulin resistance (IR) and hyperglycaemia despite reduced visceral fat. The mechanisms underlying both high phosphate and high glucose levels (which have been epidemiologically linked to increased morbidity and mortality in the general population [15, 16]) in acromegaly require further clarification.

In this study, we determined the levels of serum sKlotho in patients with acromegaly before and after transsphenoidal resection of the GH-producing pituitary adenoma.

Patients and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Patients

Twenty-nine consecutive patients with newly diagnosed acromegaly, referred to the University Hospital Zurich between 2006 and 2009, were considered eligible for this study. Of these 29 patients, 24 (nine women/15 men, age 28–76 years) were included in this prospective case series. All participants provided written informed consent, and the study was approved by the local ethics committee. Diagnosis of acromegaly was based on typical clinical findings, elevated IGF-I concentrations and nonsuppressible GH during a 75-g oral glucose tolerance test. The GH excess was caused by a pituitary adenoma in all patients. Adenoma size (in mm3) estimation and details of the surgical procedure have been reported elsewhere [17]. The initial tumour size ranged from 50 to 5091 mm3 (median 1844 mm3). All patients underwent intraoperative magnetic resonance imaging-guided transsphenoidal surgery as the primary treatment, performed by the same surgeon. Clinical examination and laboratory investigations were performed before surgery and at a postoperative follow-up visit 22–124 (median 86) days after surgery.

Biochemical analyses

Blood samples were drawn after overnight fasting. IGF-1 was determined by radioimmunoassay (RIA) [18]. Acid labile subunit (ALS) was measured by ELISA (Mediagnost, Reutlingen Germany), and GH by immunoradiometric assay (hGH-RIACT; CIS Bio International, Oris Industries, Gif-Sur-Yvette, France). Carboxy terminal FGF23 (Immutopics, San Clemente, CA, USA) and human sKlotho (Immuno-Biological Laboratories Co., Ltd. Japan) were measured by ELISA according to the manufacturers’ instructions. Fasting plasma glucose was measured with a Roche-Hitachi Modular Clinical Chemistry analyser using commercial tests from Roche Diagnostics (Rotkreuz, Switzerland). Insulin was determined using a solid-phase RIA (CIS Bio international, Oris Industries). The homoeostasis assessment model was used to estimate IR [19]. Creatinine was measured by the kinetic Jaffe reaction (Hitachi P-Modular system, Roche Diagnostics). Estimated GFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula, which seems to be more accurate than the widely used Modification of Diet in Renal Disease (MDRD) Study equation, especially at estimated GFR >60 mL min−1 1.73 m2 [20].

Statistical analysis

Data are presented as means ± SEM. Values before and after surgery were compared using the Wilcoxon signed-rank test. Correlation coefficients between Klotho and other variables were assessed using Spearman’s correlation formula. Multiple regression analysis was performed with Klotho as the dependent variable, and age, gender and levels of GH, phosphate and creatinine as independent variables. Statistical analyses were conducted using Stata SE10 software (StataCorp, College Station, TX, USA). P values < 0.05 were considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Surgery resulted in decreased clinical disease activity as judged by all patients and their physicians. Body weight declined from 87.1 ± 4.2 to 85.5 ± 4.2 kg (= 0.02), and body mass index from 28.3 ± 1.0 to 27.9 ± 1.0 kg m2−1 (= 0.03). After surgery, nadir GH levels of <1 μg L−1 either in a random blood sample or during the oral glucose tolerance test were observed in 19 of the 24 patients. In the remaining five patients, postoperative nadir GH levels remained >1 μg L−1, indicating residual disease activity. The effects of transsphenoidal surgery on biochemical markers are summarized in Table 1.

Table 1. Biochemical parameters before and after surgery
 Before surgery (mean ± SEM)After surgery (mean ± SEM)Before versus after P values
  1. aHomoeostasis model assessment.

  2. bEstimated GFR using CKD-EPI formula.

  3. cRelative units.

  4. dFor comparison, mean values (± SD) in 26 healthy controls: 0.6 ± 0.2 ng mL−1.

GH (μg L−1)26.3 ± 5.22.6 ± 0.6<0.0001
IGF-1 (μg L−1)588 ± 35193 ± 12<0.0001
ALS (mU mL−1)2638 ± 1541316 ± 67<0.0001
Fasting plasma glucose (mmol L−1)5.8 ± 0.25.2 ± 0.1<0.001
Insulin (pmol L−1)211.6 ± 44.1100.4 ± 9.8<0.0001
HOMAa7.5 ± 0.93.3 ± 0.4<0.001
Creatinine (μmol L−1)71 ± 380 ± 3<0.001
eGFR (mL min−1)b99 ± 391 ± 3<0.01
Phosphate (mmol L−1)1.37 ± 0.041.06 ± 0.02<0.001
FGF23 (RUc mL−1)21.3 ± 2.717.2 ± 2.20.03
Klotho (ng mL−1)d4.2 ± 0.70.7 ± 0.1<0.0001

GH, IGF-1 and ALS

In all 24 patients, transsphenoidal resection of the GH-producing adenoma resulted in decreased GH, IGF-1 and ALS levels (Table 1 and Fig. 1; < 0.0001 for all parameters).

image

Figure 1. Levels of (a) IGF-1, (b) ALS and (c) sKlotho before and after surgery. Patients with suppressible (circles, solid lines) and non-suppressible GH (triangles, dashed lines) after surgery.

Download figure to PowerPoint

Glucose homoeostasis

Transsphenoidal surgery improved plasma glucose and insulin sensitivity as shown in Table 1. Three patients with type 2 diabetes before surgery (all receiving insulin therapy) were excluded from analysis of glucose homoeostasis. Postoperatively, insulin was replaced by oral antidiabetic treatment in two patients, and the other patient had normal glucose levels without medication.

Renal function, phosphate and FGF23

Postoperatively, creatinine levels increased and estimated GFR decreased accordingly. Phosphate and FGF23 levels decreased concomitantly in most patients. All these changes were statistically significant.

Klotho

sKlotho levels were high preoperatively (4.2 ± 0.6 ng mL−1), on average corresponding to approximately four times the upper limit of normal. To date, a reliable reference range has not been established; therefore, we measured sKlotho in serum samples obtained from 26 healthy volunteers (11 women/15 men, mean age ± SD 39 ± 8 years) as a control: values ranged from 0.3 to 1.3 ng mL−1 (mean ± SD 0.6 ± 0.2 ng mL−1). Before surgical treatment of acromegaly, sKlotho correlated negatively with age (r = −0.45, = 0.02), and positively with GH (r = 0.64, = 0.0007) (Fig. 2), IGF-1 (r = 0.57, P =0.003) and estimated tumour size (r = 0.5, = 0.01). In multiple regression analysis, sKlotho was associated with GH after correction for age, gender and levels of phosphate and creatinine (= 0.029; beta-regression coefficient 0.056, 95% confidence interval 0.01–0.10). After surgery, the level of sKlotho decreased in all 24 patients; comparable decreases were also observed in the level of IGF-1 (Fig. 1 and Table 1). sKlotho and IGF-1 were similarly correlated with GH (r = 0.66 and r = 0.63, respectively, both < 0.001) and more strongly associated than ALS and GH (r = 0.11, not significant).

image

Figure 2. Correlation between sKlotho and GH before surgery, r = 0.64, P = 0.0007.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

To the best of our knowledge, this is the first reported study of an acquired disease in humans with dramatically elevated levels of sKlotho. In our acromegalic patients, sKlotho levels were markedly increased in relation to GH excess and returned towards normal after transsphenoidal resection of the GH-producing adenoma. Reversal of Klotho levels after surgery indicates a causal role of the GH-producing adenoma. It is remarkable that sKlotho (presumably kidney-derived) depends on GH to a comparable extent as well-known IGF-1 and more distinctively than (liver-derived) ALS.

From an endocrinological point of view, acromegaly is characterized by an excess of the triad of anabolic peptide hormones: GH, IGF 1 and insulin. It has unique biochemical and endocrine features, distinct from those of more common conditions such as renal failure and type 2 diabetes: high phosphate levels cannot be attributed to decreased GFR, and IR and hyperglycaemia are not caused by increased visceral fat. Although levels of phosphate and FGF23 [21] are high in acromegaly, calcitriol [13] is also increased in this disease. FGF23 is a phosphaturic hormone produced by osteocytes, which inhibits renal phosphate reabsorption and calcitriol production, and normally increases with decreasing GFR. Of note, serum phosphate is high despite increased GFR and high levels of FGF23 in acromegaly [21], indicating a state of renal FGF23 resistance. Currently, it is uncertain whether changes in Klotho turnover account for the high threshold of renal phosphate reabsorption and FGF23 resistance with GH excess, and the mechanisms by which acromegaly leads to excess sKlotho remain unclear. Secreted sKlotho may arise either from a distinct transcript or from enzymatic clipping of membrane Klotho [22–24], and whether high sKlotho levels also reflect an abundance of renal membrane Klotho in acromegaly remains to be determined. It seems that FGF23 resistance cannot be explained by high sKlotho per se, unless it is assumed that sKlotho is elevated at the expense of membrane Klotho, which serves as a coreceptor for FGF23. We hypothesize that enhanced clipping of the extracellular Klotho domain may result in excess sKlotho contributing to IR and deficient membrane-bound Klotho with FGF23 resistance.

It should be mentioned that we did not assess further parameters related to bone (FGF23 production site) metabolism or calcium and phosphate homoeostasis (potential targets of Klotho action), such as serum and urinary calcium, calcitriol, parathyroid hormone or maximal renal tubular phosphate reabsorption.

With regard to life expectancy, apparent contradictions concerning the action of GH should be considered. GH is generally thought to be associated with positive attributes such as growth, health, youth, muscle mass and well-being, yet the bulk of the scientific evidence suggests that signalling through GH, IGF-1 and insulin receptors is related to a shortened lifespan in adults [25]. Our clinical data do not provide information as to whether high sKlotho levels do indeed attenuate insulin and IGF-1 receptor signalling in patients with acromegaly, but it is clear that excess Klotho does not predict longevity in these patients.

In conclusion, we have shown that sKlotho levels are markedly elevated in relation to GH excess in patients with acromegaly. Levels of sKlotho and IGF-1 are similarly dependent on GH excess. High sKlotho concentrations were reversed towards normal levels after transsphenoidal resection of the GH-producing adenoma.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

We thank Samuel Henz for assistance with the statistical calculations.

Conflict of interest statement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

The authors have no conflicts of interest to declare.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References