Homocysteine Metabolism in Children with Idiopathic Nephrotic Syndrome

Abstract

Background

Homocysteine metabolism is altered in children with idiopathic nephrotic syndrome. Hyperhomocysteinemia is a risk factor of early atherosclerosis and glomerulosclerosis and may occur at time of first occurrence of idiopathic nephrotic syndrome.

Methods

Thirty children with first episode of idiopathic nephrotic syndrome (FENS) aged 1–16 years along with 30 age‐ and sex‐matched healthy controls were enrolled in this study. Homocysteine and cysteine were measured with HPLC; vitamin B₁₂ and folic acid were measured with electro‐chemilumiscence immunoassay. Primary outcome measure was plasma homocysteine level in children with FENS and in controls. Secondary outcome measures were (1) plasma and urine homocysteine and cysteine levels in children with FENS at 12 weeks and 1 year (remission) and (2) plasma and urine levels of vitamin B₁₂ and folic acid in children with FENS, at 12 weeks and 1 year (remission).

Results

Plasma homocysteine and cysteine levels were comparable to controls in children with FENS, at 12 weeks and 1‐year remission. Plasma levels of vitamin B12 and folic acid were significantly decreased compared to controls in FENS due to increased urinary excretion, which normalize during remission at 12 weeks and 1 year. Urinary homocysteine and cysteine levels were significantly raised in FENS compared to controls and continued to be raised even at 12‐week and 1‐year remission.

Conclusion

Homocysteine metabolism is deranged in children with FENS. Renal effects of long‐term raised urinary homocysteine levels need to be studied.

Introduction

Idiopathic nephrotic syndrome (INS) is the most common glomerulopathy in childhood with an annual incidence of 20–40 per million population all over world.1 Patients with INS are at risk of many complications like infections, hypovolemic shock, hypertension, corticosteroid side effects, and thrombosis. The incidence of thrombo‐embolic complication in nephrotic syndrome patients is approximately 3%. Venous thrombosis is threefold more frequent compared to arterial thrombosis.2 The increase in levels of homocysteine (Hcys) has been an independent risk factor for both arterial and venous thrombosis.3 There are long‐term studies in adult nephrotic population, indicating association of cardiovascular diseases in these patients.4 Risk factors for early atherosclerosis in the adult population include hyperhomocystenemia (hHcy), proteins of chronic inflammatory process, pro‐thrombotic factors, and adhesion molecules.5 Various studies in the adult population have linked these factors to endothelial dysfunction, and the complications of subsequent atherothrombosis. Many of these risk factors appear in pediatric patients at the time of first occurrence or during aggravation of INS.6 Their appearance is related to excessive loss of albumin in urine (leading to hypoproteinemia, compensatory dysproteinemia and hyperlipidemia) or to steroid therapy (leading to obesity, glucose utilisation disorders, and increased arterial hypertension). They may also predispose the pediatric patients to endothelial dysfunction. Recent studies have demonstrated that hHcy resulted in glomerular dysfunction and glomerular sclerosis independently of the elevations of arterial pressure, indicating that hHcys may be one of the important independent pathogenic factors resulting in glomerular injury in hypertension.7

Several studies have been done to find the levels of homocysteine in patients with nephrotic syndrome; however the results are controversial. Some studies showed increased6, 8, 9 levels of homocysteine in cases of nephrotic syndrome and some with decreased,10 while others reported no change in homocysteine levels compared to controls.11 We hypothesized that homocysteine metabolism is altered in children with idiopathic nephrotic syndrome. The objective of our study was to measure the plasma and urine homocysteine levels in children with first episode of idiopathic nephrotic syndrome (FENS) at onset and in drug‐induced remission.

Subjects and methods

Study design and patients

This was a cross‐sectional study with longitudinal follow‐up, conducted at a tertiary care teaching hospital in New Delhi, India from March 2011 to October 2012 and was approved by the Institutional Review Board (No.15/2010). Written informed consent was obtained from the legal guardian of each child before enrolment. Children aged between 1 year and 16 years with FENS were eligible for inclusion in the study. Those with secondary nephrotic syndrome, thromboembolic complications, bleeding diathesis, preexisting hypertension, diabetes mellitus, blood transfusion, drugs affecting endothelial functions, consumption of vitamin B₁₂ and folic acid prior to sampling, serum creatinine >1.0 mg/dL and refusal of informed consent were excluded from the study. Sample size was calculated with power of 0.9, alfa error of 0.05 and standard deviation of 7.7 in the study group and 1.6 in the control group.6 Thirty children with FENS and 30 age‐ and sex‐matched normal healthy children served as controls for the study. Clinical assessment of each participant included a review of medical records with respect to history, blood pressure measurement, height, weight, urine analysis, and quantification of proteinuria, serum electrolytes, blood urea, serum creatinine, and other relevant investigations. Children were followed up in the pediatric nephrology clinic for 1 year. FENS was treated with 2 mg/kg of prednisolone for 6 weeks followed by 1.5 mg/kg for next 6 weeks. Frequent relapse nephrotic syndrome (FRNS) was defined as two or more relapses within 6 months of the initial episode or more than three relapses within any 12‐month period. Steroid‐dependent nephrotic syndrome (SDNS) was defined as two consecutive relapses when on alternate‐day prednisolone or within 14 days of discontinuation of prednisolone. Steroid resistance nephrotic syndrome (SRNS) was defined as the absence of remission despite therapy with daily prednisolone at a dose of 2 mg/kg/day for 4 weeks.12Ultrasound‐guided percutaneous renal biopsy was performed in all cases of SRNS. Primary outcome measure was plasma homocysteine level in children with FENS and in controls. Secondary outcome measures were (1) plasma and urine homocysteine and cysteine levels in children with FENS at 12 weeks and 1 year (remission). (2) Vitamin B₁₂ and folic acid levels in plasma and urine in children with FENS at 12 weeks and 1 year (remission).

Sample collection and storage

The blood sample was collected by venipuncture into a vecutainer tube containing EDTA. Plasma obtained by centrifuging the blood at 1,300 rpm for 15 minutes at 4°C. Plasma samples were stored at –80°C until analyzed. The urine samples were collected from patients at the hospital. After collection, crude samples were filtered and immediately centrifuged at 2,500 rpm for 10 minutes. After centrifugation supernatant were collected and stored at –80°C in 1 mL cryo tubes.

Homocysteine and cysteine measurement in plasma

Plasma was treated with sodium borohydride to reduce disulfide bonds. The borohydride initially prepared in an alkaline solution and then activated by addition of acid causes effervescence. Amyl alcohol controls foaming. Borohydride reduction of disulfide bonds releases homocysteine from its bonding proteins. The thiol group of free homocysteine or cysteine was then derivatized with fluorescent adduct monobromobimane to form fluorescent homocysteine–bimane conjugate. Proteins were removed by acid precipitation with perchloric acid and centrifugation. Supernatant was neutralized with Tris and then injected into high‐performance liquid chromatography (HPLC). Homocysteine–bimane or cysteine–bimane was resolved on a C‐18 reverse phase column and detected fluorometrically. Results were then quantified by taking the area for the homocysteine–bimane peak or cysteine–bimane peak and calculating its concentration using a regression equation derived from a standard curve.13, 14, 15, 16

Homocysteine and cysteine measurement in urine

This was done by standard HPLC essay. Undiluted urine was derivatized by adding 30 μL sample with 30 μL of 4 mol/L NaBH₄ in 0.066 mol/L NaOH and 333 mol/L DMSO (dimethylsulfoxide), 10 μL of 2 mmol/L EDTA and 1.65 mmol/L dithioerythritol (DTE), 10 μL of 1‐octanol, and 20 μL of 1.8 mol/L HCL. After 3 minutes (to allow reduction of disulfides), 100 μL of 1.5 mol/L ethylmorpholine buffer, 400 μL of H₂O, and 20 μL of 0.025 mol/L monobromobimane were added. The derivatization (with monobromobimane) terminated after 3 minutes by adding 40 μL of glacial acetic acid.14

Twenty microliters of derivatized sample was injected into a 150×4.6 mm ODS hypersil column, equilibrated with 30 mmol/L ammonium nitrate and 40 mmol/L ammonium formate buffer, pH 3.50. Thiols eluted from the column with the linear gradient of acetonitrile (from 0% to 10% in 11 minute). A flow rate of 2 mL/min and ambient temperature were used.

Vitamin B₁₂ measurement

The ElecsysVitamin B₁₂ is an electro‐chemilumiscence immunoassay (ECLIA) which employs a competitive test principle using the intrinsic factor specific for B₁₂. B₁₂ in the sample competes with the added B₁₂ labeled with biotin for the binding sites on ruthenium labeled intrinsic factor complex (labeled with ruthenium complex Tris (2,2′‐bipyridyl) ruthenium (II)‐complex (Ru (bpy) ₃ ²⁺).17, 18

Folate measurement

The elecsysfolate is an ECLIA essay that employs a competitive test principle using natural folate binding protein (FBP) specific for folate. Folate in the sample competes with the added folate (labeled with biotin) for the binding sites on the FBP (labeled with ruthenium complex Tris (2,2′‐bipyridyl) ruthenium (II)‐complex (Ru (bpy) ₃ ²⁺).19, 20

Statistical analysis

Statistical analysis was carried out using STATA ver. 9.0 (STATA Corp, College Station, TX). Continuous variables were presented as the mean± standard deviation (SD) or median (minimum–maximum) while categorical variables were presented as the frequency and percentage. For continuous variables, a p value was calculated by a t‐test and Wilcoxon rank‐sum test, while categorical variables were tested by Fisher's exact test. A p value of <0.05 was considered to be significant.

Results

Baseline characteristics in children with FENS were similar to those of the controls except for blood pressure which was higher in the former category (Table 1). Children with nephrotic syndrome had low serum albumin and total proteins. They also had high serum cholesterol, serum LDL, serum triglycerides, serum VLDL, urinary albumin, and Up/Uc ratio (p < 0.001). The levels of serum sodium, potassium, creatinine, and eGFR were same compared to controls. Of the study group (n = 30), four turned into SRNS (initial resistance) and two patients developed FRNS at 1 year of follow‐up.

Table 1.

A. Demographic profile of patients and controls. B. Base line investigations of patients and controls

Parameter	INS (n = 30)	Controls (n = 30)	p value
Age (years)	7.36 ± 3.70	8.89 ± 2.76	0.07
Height (cm)	117.48 ± 21.58	122.66 ± 16.62	0.30
Weight (kg)	23.06 ± 9.62	25.76 ± 9.18	0.27
BMI	16.11 ± 2.55	16.66 ± 3.68	0.50
Systolic BP (mmHg)	106.86 ± 6.74	102.76 ± 4.53	*0.001
Diastolic BP (mmHg)	68.9 ± 6.09	64.20 ± 3.8	*0.001
Parameters	INS (n = 30)	Controls (n = 30)	p value
Hemoglobin (g/dL)	11.28 ± 1.42	11.54 ± 0.97	0.41
Creatinine (mg/dL)	0.61± 0.19	0.67 ± 0.13	0.15
Serum sodium (mEq/L)	138.23± 5.32	138.83 ± 3.76	0.61
Serum potassium (mEq/L)	4.41 ± 0.67	4.14 ± 0.41	0.06
Serum albumin (g/dL)	1.80 ± 0.30	4.10 ± 0.30	0.000
Serum cholesterol (mg/dL)	422.96 ± 88.91	130.93 ± 31.58	*0.001
Up/Uc ratio (mg/mg)	13.7(2.1–61.6)	0.2 (0.12–0.5)	*0.001

*p value significant.

Φ‐test, = unpaired t‐test; cm = centimeters; kg = kilograms; mg = milligrams; g = grams.

Plasma homocysteine and cysteine levels were similar in FENS, at 12‐week and 1‐year remission, respectively, with no significant difference in the levels during follow‐up (p value = 0.60 for homocysteine and 0.93 for cysteine).There was no change in plasma homocysteine and levels compared to control levels at FENS, at 12‐week and 1‐year remission, respectively (Table 2A). The homocysteine and cysteine levels in urine were found to be higher in FENS, at 12‐week and 1‐year remission compared to controls. There was no change in urinary homocysteine and cysteine values from initial episode, at 12‐week and 1‐year remission (p values = 0.69 and 0.28, respectively) (Table 2B). Plasma vitamin B₁₂ and folate levels were significantly lower in FENS compared to controls. On achieving remission, levels of both became comparable to controls. There was a significant increase in plasma vitamin B₁₂ and folate levels between FENS, to remission at 12 weeks and 1‐year levels (p value < 0.001), but no significant change in 12‐week remission compared to 1‐year remission (p value = 0.98) (Table 3A). The urinary levels of plasma vitamin B₁₂ and folate levels were significantly higher in FENS, at 12‐week and 1‐year remission compared to control levels (Table 3B). There was no correlation between the urinary protein excretion and the urinary homocysteine and cysteine levels. Serum cholesterol values correlated with urinary cysteine levels (Table 4, Figure 1).

Table 2.

(A) Plasma homocysteine and cysteine levels in FENS and remission. (B) Urine homocysteine and cysteine levels in FENS and remission

A.	Homocysteine (μmol/L)	Cysteine (μmol/L)
Control	7.98 (3.91–11.61)	145.28 ± 35.86
INS (first episode)	6.46 (3.06–24.4)	138.5 ± 45
INS (12‐week remission)	6.4 (2.30–16.48)	136.3 ± 42.36
INS (1‐year remission)	7.04 (2.96–17.13)	133.35 ± 41.10
B.	Homocysteine (μmol/L)	Cysteine (μmol/L)
Control	6.42 (3.29–8.78)	25.89 (14.43–64.10)
INS (first episode)	*10.58 (1.86–28.84)	*70.17 (12.59–202.07)
INS (12‐week remission)	*10.16 (3.72–29.99)	*54.86 (7.14–94.27)
INS (1‐year remission)	*9.54 (3.16–30.1)	*59.66 (17.98—91.10)

*p value significant.

ϕ‐test = Mann–Whitney test; μmol/L = micromoles per liter.

Table 3.

(A) Plasma vitamin B₁₂ and folate levels in FENS and remission. (B) Urine vitamin B₁₂ and folate levels in FENS and remission

A.	Vitamin B12 (pg/mL)	Folate (ng/mL)
Control	280.4 (103.9–654.8)	5.01 (1.63–18.98)
INS (first episode)	*167.9 (30.92–603.5)	*2.96 (0.64–7.7)
INS (12‐week remission)	214.2(53.74–311.40)	5.6 (1.14–9.79)
INS (1‐year remission)	280.4 (103.9–654.8)	6.07 (2.0–18.0)
B.	Vitamin B12 (pg/mL)	Folate (ng/mL)
Control	30.00 (30.0–30.0)	0.64 (0.64–0.64)
INS (first episode)	*74.98 (30–300.5)	*3.94 (0.64–20.0)
INS (12‐week remission)	*49.6 (32.0– 97.6)	*3.5 (0.64–10.92)
INS (1‐year remission)	*67.83 (32–111.2)	*5.04 (0.64–13.55)

pg/mL = picograms per milliliter; ng/mL = nanograms per milliliter.

*p value significant; ϕ‐test, Mann–Whitney test and Friedman test.

Table 4.

Coefficients of correlation between homocysteine, cysteine, vitamin B₁₂, folate and serum cholesterol, serum albumin, proteinuria in patients with the nephrotic syndrome

	Serum cholesterol	Plasma albumin	Up:Uc ratio
Homocysteine	r = 0.25 p = 0.17	r = 0.22 p = 0.23	r = 0.01 p = 0.94
	rp = 0.17	rp = 0.23	rp = 0.94
Cysteine	r = 0.48	r = 0.18	r = 0.24
	* p = 0.006	p = 0.32	p = 0.20
Vitamin B12	r = 0.30	r = 0.23	r = 0.15
	p = 0.10	p = 0.20	p = 0.41
Folate	r = 0.05	r = –0.04	r = 0.04
	p = 0.77	p = 0.79	p = 0.8

*p value significant; Φ, Spearman's correlation coefficient.

Figure 1.

Scattergraph depicting the relationship of the serum cholesterol to urine cysteine.

Discussion

Our study shows that children with FENS have plasma Hcys levels comparable to controls; however, this was associated with the marked increase in urinary Hcys excretion. This is plausible as most of the Hcys in the plasma is bound to albumin.13 There was concurrent increased urinary excretion of vitamin B₁₂ and folate which normalized during remission. Elevation of urinary Hcys can be explained by the presence of albumin bound homocysteine in urine. Children with FENS had low plasma levels of vitamin B₁₂ and folate, a finding previously reported by Podda et al.8 Low plasma vitamin B₁₂ and folate were because of increased urinary loss of these vitamins in the urine due to proteinuria. Normal plasma Hcys in children with FENS should not be necessarily viewed as a favorable marker. This is because the normal or reduced plasma Hcys in INS is a result of diminished albumin bound Hcys as opposed to the free Hcys fraction. Conversely free Hcys levels may actually be elevated in children with INS, as reported by Tkaczyk et al.21 Free Hcys species inactivate NO, promote the generation of oxygen‐derived free radicals, induce tissue factor release, and cause endothelial cell injury.22 Hcys is converted to methionine by remethylation and to cysteine via trans‐sulfuration. Our study demonstrated raised urinary levels of cysteine in children with FENS, pointing toward activation of trans‐sulfuration pathway. Potent toxicity of excess cysteine has been recognized as an increased cardiovascular risk factor.23

The major strength of our study was the demonstration that despite remission at 12‐week and 1‐year follow‐ups, children with INS continued to have increased urinary excretion of homocysteine and cysteine compared to controls. Long‐term effects of these findings on renal tissue need to be analyzed, especially in children with FRNS/SDNS and SRNS who have protracted course. Tenderenda et al. recently reported increased urinary levels of Hcys in FRNS/SDNS children even during remission.24 Li et al. demonstrated that increased plasma Hcys may directly act on the glomeruli to produce sclerotic changes.7 On the basis of these results, they proposed that elevations of plasma Hcys levels contribute at to glomerular damage or sclerosis in rats. Similarly Chen et al. demonstrated that chronic elevations of the plasma Hcys concentrations in rats resulted in arteriosclerotic changes and glomerular dysfunction and sclerosis accompanied by a sustained low level of plasma adenosine.25 It is possible that the sclerotic effect of hHcys is associated with decreased adenosine. Indeed, it was demonstrated that decreased adenosine levels are associated with enhanced proliferation or growth of vascular smooth muscle cells and sclerotic changes in arteries or glomeruli.26

Limitations of our study include single‐center, homogenous population and small sample size. In conclusion Hcys metabolism is altered in children with FENS. Renal effects of long‐term raised urinary homocysteine levels needs to be studied.

Conflict of Interest

All the authors have declared no competing interest.

Funding

None.