Lower urinary α‐Klotho is associated with lower angiotensin‐(1‐7) and higher blood pressure in young adults born preterm with very low birthweight

Abstract

Early‐life factors including preterm birth and VLBW increase the risk of hypertension, but the mechanisms remain poorly understood. Reductions in the anti‐aging protein α‐klotho are associated with hypertension, possibly due to angiotensin (Ang) II activation, but the mechanisms are incompletely understood and clinical evidence is lacking. The association of α‐klotho with the alternative Ang‐(1‐7) pathway, which counteracts Ang II to lower BP, is undescribed. We hypothesized that lower urinary α‐klotho is associated with higher BP and lower urinary Ang‐(1‐7) in preterm‐born VLBW young adults. In a cross‐sectional analysis of data from a prospective cohort of 141 preterm‐born VLBW young adults, we assessed the associations among urinary α‐klotho/creatinine, Ang II/creatinine, Ang‐(1‐7)/creatinine, Ang II/Ang‐(1‐7), and BP using generalized linear models adjusted for age and hypertensive pregnancy and conducted a sensitivity analysis in 32 term‐born young adults. Among those born preterm, lower α‐klotho/creatinine was associated with higher systolic BP (adjusted β (aβ): −2.58 mm Hg, 95% CI −4.99 to −0.17), lower Ang‐(1‐7)/creatinine (ln aβ: 0.1, 0.04‐0.16), and higher Ang II/Ang‐(1‐7) (ln aβ: −0.14, −0.21 to −0.07). In term‐born participants, α‐klotho/creatinine was inversely associated with Ang II/creatinine (ln aβ: −0.15, −0.27 to −0.03) and Ang II/Ang‐(1‐7) (ln aβ: −0.15, −0.27 to −0.03). In preterm‐born young adults with VLBW, lower urinary α‐klotho/creatinine was associated with higher SBP, lower urinary Ang‐(1‐7)/creatinine, and higher urinary Ang II/Ang‐(1‐7). Reduced renal α‐klotho expression could lead to renal Ang‐(1‐7) suppression as a novel mechanism for the development of hypertension among individuals born preterm with VLBW.

1. INTRODUCTION

Cardiovascular disease remains the leading cause of mortality worldwide, and among all cardiovascular risk factors, hypertension is the leading cause of mortality and the second leading cause of kidney failure, now affecting approximately 46% of US adults. ¹ , ² , ³ , ⁴ Despite this disease burden, the underlying mechanisms contributing to the development of hypertension and cardiovascular disease remain incompletely understood, especially the antecedent pathophysiologic alterations which occur early in the lifespan. The anti‐aging protein α‐klotho, the renal co‐receptor of fibroblast growth factor 23, has been shown to be inversely associated with cardiovascular disease and hypertension, but the mechanisms responsible for α‐klotho's putative role are not known. ⁵ , ⁶ , ⁷ , ⁸ , ⁹ Experimental and clinical studies suggest that reduced α‐klotho may contribute to increased BP via upregulation of the ACE‐angiotensin (Ang) II‐Ang II type 1 receptor pathway within the renin‐Ang‐aldosterone system (RAAS), leading to increased renal sodium retention, vasoconstriction, and oxidative stress. ¹⁰ , ¹¹ , ¹² The ACE2‐Ang‐(1‐7)‐Mas receptor RAAS pathway, in part, attenuates Ang II's effects on BP via increased sodium excretion and vasodilation and reduced oxidative stress, but the association of α‐klotho with Ang‐(1‐7) has not been described. ¹³ , ¹⁴

Premature birth and associated adverse perinatal exposures such as VLBW are increasing worldwide. ¹⁵ Prematurity is an emerging risk factor for hypertension and cardiovascular disease, but the underlying mechanisms remain undefined. ¹⁶ Neonates born preterm have lower plasma soluble α‐klotho levels compared to those born at term, but the long‐term association of preterm birth with α‐klotho, particularly in the urine, is unknown. ¹⁷ We previously showed that preterm birth and other perinatal risk factors are associated with lower plasma and urine Ang‐(1‐7) concentrations relative to Ang II [ie, higher Ang II/Ang‐(1‐7)], in addition to higher BP. ¹⁸ , ¹⁹ , ²⁰ Thus, we proposed that reduced urinary α‐klotho coupled with reduced Ang‐(1‐7) in the urine may serve as a mechanism for the development of hypertension in young adults born preterm with VLBW. We hypothesized that (a) lower urinary soluble α‐klotho is associated with higher BP and higher urinary Ang II/Ang‐(1‐7); (b) higher urinary Ang II/Ang‐(1‐7) is associated with higher BP; and (c) the magnitude of these associations will be greater in young adults born preterm with VLBW compared to those born at term.

2. METHODS

2.1. Study participants and design

Detailed study methods have been previously published. ¹⁸ Participants were part of an ongoing prospective birth cohort (Prenatal Events‐Postnatal Consequences: PEPC) of 193 individuals born preterm with VLBW between January 1, 1992, and June 30, 1996, at a regional perinatal center (Forsyth Medical Center, Winston Salem, NC). Participants were first evaluated at age 14 years (PEPC1) and again as young adults (PEPC2) aged 18‐23 years (mean age 19.9 years). In the current analysis, we reported data obtained from participants at the second of two PEPC2 study visits, when laboratory samples were obtained. Inclusion criteria for the preterm‐birth cohort were singleton birth, gestational age <37 weeks, birthweight <1500 g, and available clinical data through 1 year corrected gestational age. Term‐born participants were recruited at age 14 years by newspaper advertisements and word of mouth; inclusion criteria were birth at the same center over the same time period, gestational age >37 weeks, birthweight >2500 g, and no antenatal steroid exposure. Exclusion criteria for both groups were presence of congenital anomalies or major genetic syndromes, or being a ward of the state. The Wake Forest School of Medicine and Forsyth Medical Center IRBs approved the study. Participants provided written informed consent and were compensated.

2.2. Data collection

To obtain maternal and participant birth characteristics, a study nurse reviewed participants' medical records and research databases, and participants and their parents/guardians completed questionnaires. For the current analysis, the presence of maternal hypertensive pregnancy was determined via review of medical records for preterm‐born participants and via questionnaire for term‐born participants. One preterm‐birth participant was missing maternal hypertensive pregnancy information from the medical record and was assigned the value recorded from the questionnaire (kappa = 0.78 for agreement between the medical record and questionnaire in the preterm‐birth cohort). The authors classified participants as SGA if their birthweight was < 10th percentile for gestational age and sex. ²¹ participant demographics and characteristics as young adults were recorded, including participant‐reported race (black or non‐black). The authors measured height and weight and calculated BMI; obesity was defined as BMI ≥ 30.0 kg/m². ²² Study personnel were blinded to participants' preterm birth status when assessing outcomes and analyzing laboratory samples.

Trained research staff measured BP in accordance with established guidelines using a mercury manometer and an appropriately sized BP cuff. ²³ Participants were seated quietly for at least 5 minutes with full arm support. At each of the two PEPC2 study visits, three BP measurements were obtained with at least 1 minute between each measurement, and the averages of both the three systolic and diastolic BP (SBP, DBP) measurements were recorded. The authors reported the average SBP and DBP measured at the second study visit, when urine samples were obtained. The authors also reported the prevalence of high BP (elevated BP or hypertension, see below) based on the average measurements obtained at both PEPC2 study visits per the 2017 Adult High Blood Pressure Clinical Practice Guidelines: high BP if SBP or DBP ≥ 120/80 mm Hg, further classified as (a) elevated BP if 120‐129/<80 mm Hg; (b) hypertension if ≥130/80 mm Hg; (c) stage 1 hypertension if 130 to 139/80‐89 mm Hg; and (d) stage 2 hypertension if ≥140/90 mm Hg. ⁴

2.3. Laboratory measurements

Detailed laboratory methods including measurements of the RAAS have been published previously. ¹⁸ Briefly, spot urine samples were collected in an HCl‐containing tube to measure soluble α‐klotho, Ang II, and Ang‐(1‐7). α‐Klotho was determined by ELISA (Immuno‐Biological Laboratories) according to the manufacturer's recommendations with modification, including expansion of the standard curve at lower concentrations to improve delineation of values at lower α‐klotho levels (sensitivity 6.15 pg/mL). The peptides Ang II and Ang‐(1‐7) were measured in the urine with separate RIAs previously validated against mass spectrometry. ²⁴ Blood was collected in the seated position; the authors measured serum and urine creatinine using a modified Jaffe assay traceable to isotope dilution mass spectrometry and serum cystatin C using an ELISA (R&D Systems, MN). The authors estimated the GFR using the CKD‐EPI creatinine‐cystatin C equation (2012). ²⁵ The authors corrected the urinary concentrations of α‐klotho, Ang II, and Ang‐(1‐7) for urine creatinine, and the authors calculated the ratio of Ang II to Ang‐(1‐7) in the urine. Sample results below the lower limit of detection were assigned the value of the lower limit of detection divided by the square root of two, a well‐validated technique that the authors previously used in this cohort. ¹⁸ , ²⁶

2.4. Statistical analyses

The current analyses are secondary aims of the main study. The authors examined measures of central tendency and dispersion and frequencies. The authors used t test, Wilcoxon rank‐sum test, chi‐square test, and Fisher's exact test for between‐group comparisons. The authors used Pearson and Spearman coefficients to assess for correlations between continuous variables. The authors improved the distributional characteristics of the variables with ln transformation if indicated. A two‐sided alpha level <0.05 was considered statistically significant. The authors utilized generalized linear models (identity link, normal distribution) to estimate the associations of urinary α‐klotho with the urinary RAAS and with BP. As a secondary analysis, the authors estimated the association between the urinary RAAS and BP. The authors employed two sensitivity analyses: The authors excluded participants whose urinary α‐klotho concentrations were below the lower limit of detection, and The authors estimated associations in the models in the term‐born cohort and a priori evaluated for effect modification by preterm birth status by including interaction terms in the regression models. An interaction term P value <.2 was suggestive of effect modification, based on our previous work. ¹⁸ The authors used directed acyclic graphs to evaluate for potentially confounding factors based on the literature a priori and identified a minimally sufficient adjustment set for inclusion in the models that consisted of PEPC2 age and maternal hypertensive pregnancy. ¹⁷ , ²⁴ , ²⁷ The authors used Enterprise Guide software, Version 7.11 of the SAS System for Windows for all of the analyses (SAS Institute Inc).

3. RESULTS

Of the 193 preterm‐born participants who were enrolled in PEPC1, five were excluded (Figure 1). Of the 185 participants who completed PEPC1, 148 participated in PEPC2. The authors enrolled an additional 28 preterm‐born young adults who met inclusion criteria for but were not originally enrolled in PEPC1. Of the 176 preterm‐born PEPC2 participants, 35 were excluded from this analysis. Among the 52 term‐born study participants, 17 were lost to follow‐up by PEPC2 and three were further excluded.

FIGURE 1.

PEPC study flowchart

The perinatal and young adult characteristics of the preterm‐birth group compared to the term‐birth group are shown in Table 1. The preterm‐birth cohort had lower mean gestational age and birthweight, but there was no difference in the proportions of SGA. The preterm‐birth group had greater proportions of maternal hypertensive pregnancy and birth by cesarean section. As young adults, the preterm‐birth cohort was younger, shorter, and had a greater proportion of obesity. There was no difference in renal function.

TABLE 1.

Clinical and laboratory characteristics of the preterm‐ and term‐birth groups

	Preterm n = 141	Term n = 32
Perinatal
Maternal hypertensive pregnancy	54 (38.3%)*	1 (3.1%)
Cesarean section	76 (53.9%)*	7 (21.9%)
Gestational age, wk	28.0 (2.8)*	39.7 (1.2)
Birthweight, g	1067 (272)*	3499 (473)
SGA	17 (12.1%)	1 (3.1%)
Young adult
Age, y	19.9 (0.9)*	20.2 (0.7)
Female	76 (53.9%)	16 (50.0%)
Black	62 (44.0%)	11 (34.4%)
Height, cm a	164.7 (9.9)*	172.4 (8.9)
Weight, kg	72.6 (22.2)	74.0 (15.2)
BMI, kg/m2 a	26.5 (7.2)	24.8 (4.3)
Obesity a	41 (29.3%)*	3 (9.4%)
Serum creatinine, mg/dL b	0.8 (0.16)	0.81 (0.16)
Serum cystatin C, mg/L c	0.29 [0.23, 0.35]	0.31 [0.25, 0.39]
Estimated GFR, mL/min/1.73 m2 c	181.3 (28.7)	172.7 (21.1)
Systolic BP, mm Hg	110.8 (10.6)*	106.3 (9.7)
Diastolic BP, mm Hg	70.3 (9.6)	67.7 (8.1)
High BP category a	51 (36.4%)	6 (18.8%)
Elevated BP	7 (5.0%)	2 (6.3%)
Hypertension	44 (31.4%)*	4 (12.5%)
Stage 1	37 (26.4%)	4 (12.5%)
Stage 2	7 (5.0%)	0 (0.0%)
Urinary α‐klotho/creatinine, pg/g	0.11 [0.05, 0.2]	0.12 [0.07, 0.2]
Urinary Ang II/Ang‐(1‐7)	0.48 [0.38, 0.62]	0.52 [0.43, 0.71]
Urinary Ang II/creatinine, pmol/g	0.11 [0.08, 0.15]	0.11 [0.08, 0.14]
Urinary Ang‐(1‐7)/creatinine, pmol/g	0.22 [0.17, 0.28]	0.22 [0.16, 0.28]

N (%), mean (SD), median [IQR].

Abbreviation: Ang, angiotensin.

^aN = 172 (preterm 140, term 32).

^bN = 165 (preterm 133, term 32).

^cN = 159 (preterm 127, term 32).

^* P < .05 for preterm vs term comparison via chi‐square test, Fisher's exact test, or t test.

3.1. Correlation between urinary α‐Klotho and BP

α‐Klotho/creatinine was inversely correlated with SBP on bivariate analysis, though not at the level of statistical significance (P = .08; Figure 2); there was no correlation with DBP. After adjusting for age and maternal hypertensive pregnancy, lower α‐klotho/creatinine was significantly associated with higher SBP (adjusted β (aβ): −2.58 mm Hg, 95% CI −4.99 to −0.17) (Table 2). This association remained statistically significant when excluding those participants with α‐klotho concentrations below the lower limit of detection (aβ −2.61 mm Hg, 95% CI −5.1 to −0.12; n = 98).

FIGURE 2.

Urinary α‐klotho/creatinine is inversely correlated with SBP in participants born preterm with VLBW. Regression line with 95% confidence limits and Pearson correlation coefficient with corresponding P value. Cr, creatinine; SBP, systolic BP

TABLE 2.

Lower urinary α‐klotho/creatinine is associated with higher SBP, lower urinary Ang‐(1‐7)/creatinine, and higher urinary Ang II/Ang‐(1‐7) in preterm‐born young adults with VLBW

	Unadjusted	Adjusted a
SBP, mm Hg	−2.18 (−4.58 to 0.21)	−2.58 (−4.99 to −0.17)
Ang II/creatinine b	−0.02 (−0.1 to 0.06)	−0.03 (−0.11 to 0.05)
Ang‐(1‐7)/creatinine b	0.1 (0.05 to 0.16)	0.1 (0.04 to 0.16)
Ang II/Ang‐(1‐7) b	−0.12 (−0.19 to −0.06)	−0.14 (−0.21 to −0.07)

Generalized linear models with β (95% CI).

Abbreviations: Ang, angiotensin; SBP, systolic BP.

^aAdjusted for age and maternal hypertensive pregnancy.

^bα‐klotho and angiotensin peptides ln‐transformed.

3.2. Correlation between urinary α‐Klotho and the urinary RAAS

On bivariate analysis, ln α‐klotho/creatinine was significantly correlated with ln Ang‐(1‐7)/creatinine (Figure 3) and inversely correlated with ln Ang II/Ang‐(1‐7). These correlations remained significant after adjusting for age and maternal hypertensive pregnancy (aβ: 0.1, 0.04 to 0.16; and aβ: −0.14, −0.21 to −0.07, respectively) (Table 2). These associations remained statistically significant when excluding those participants with α‐klotho concentrations below the lower limit of detection (ln Ang‐(1‐7)/creatinine aβ 0.12, 95% CI 0.04 to 0.2; ln Ang II/Ang‐(1‐7) aβ −0.15, 95% CI −0.24 to −0.07; n = 98).

FIGURE 3.

Urinary α‐klotho/creatinine is correlated with urinary Ang‐(1‐7) in participants born preterm with VLBW. Regression line with 95% confidence limits and Pearson correlation coefficient with corresponding P value. Variables ln‐transformed. Ang, angiotensin; Cr, creatinine

3.3. Correlation between the urinary RAAS and BP

Our secondary analysis demonstrated that Ang‐(1‐7)/creatinine was inversely correlated with DBP (β: −15.34 mm Hg, −31.22 to 0.53, P = .06) on bivariate analysis, though it did not reach statistical significance. However, after adjusting for age and maternal hypertensive pregnancy, this association became statistically significant (aβ: −17.7 mm Hg, −33.39 to −2.01). The remaining RAAS measures were not significantly associated with SBP or DBP.

3.4. Sensitivity analysis in the term‐born cohort

Compared to the term‐birth cohort, the preterm‐birth cohort had a significantly higher mean SBP (110.8 vs 106.3 mm Hg) and a significantly greater proportion of hypertension (31.4% vs 12.5%) (Table 1). A greater proportion of the preterm‐birth group had α‐klotho levels that were below the limit of detection compared to the term‐birth group (30.5% vs 15.6%), but this was not statistically significant (P = .09). There were no preterm‐term birth differences in α‐klotho/creatinine, Ang II/creatinine, Ang‐(1‐7)/creatinine, or Ang II/Ang‐(1‐7).

On unadjusted and adjusted analyses, ln α‐klotho/creatinine was significantly and inversely associated with ln Ang II/creatinine (aβ: −0.15, −0.27 to −0.03) and with ln Ang II/Ang‐(1‐7) (aβ: −0.15, −0.27 to −0.03) (Table 3). These associations remained statistically significant when excluding those participants with α‐klotho concentrations below the lower limit of detection (ln Ang II/creatinine aβ −0.14, 95% CI −0.26 to −0.02; ln Ang II/Ang‐(1‐7) aβ −0.14, 95% CI −0.28 to −0.003; n = 27).

TABLE 3.

Lower urinary α‐klotho/creatinine is associated with higher urinary Ang II/Ang‐(1‐7) in term‐born young adults

	Unadjusted	Adjusted a
SBP, mm Hg	−1.51 (−11.99 to 8.98)	−1.63 (−12.47 to 9.21)
Ang II/creatinine b	−0.14 (−0.26 to −0.03)	−0.15 (−0.27 to −0.03)
Ang‐(1‐7)/creatinine b	0.002 (−0.14 to 0.15)	0.004 (−0.14, 0.15)
Ang II/Ang‐(1‐7) b	−0.14 (−0.26 to −0.02)	−0.15 (−0.27, −0.03)

Generalized linear models with β (95% CI).

^aAdjusted for age and maternal hypertensive pregnancy.

^bα‐klotho and RAAS ln‐transformed.

In addition, Ang‐(1‐7)/creatinine was significantly and inversely associated with DBP (aβ: −34.53, −65.31 to −3.75). An interactive effect of preterm birth on the association between α‐klotho/creatinine and Ang‐(1‐7)/creatinine was suggested but did not reach statistical significance, with a greater magnitude of effect in the preterm‐birth group (interaction term P = .17). Preterm birth did not modify the strength of the associations between α‐klotho/creatinine and SBP, Ang II, or Ang II/Ang‐(1‐7) (interaction terms P = .92, P = .23, and P = .92, respectively), nor the association between Ang‐(1‐7)/creatinine and DBP (interaction term P = .51).

4. DISCUSSION

In a cohort of young adults born preterm with VLBW, lower levels of soluble α‐klotho/creatinine in the urine were associated with higher BP, lower urinary Ang‐(1‐7)/creatinine, and higher Ang II/Ang‐(1‐7). Participants born preterm with VLBW also had higher BP and were more likely to have hypertension as compared to young adults born at term. Our results support the theory that suppression of renal α‐klotho could induce suppression of renal Ang‐(1‐7) as a potential mechanism contributing to high BP, especially in individuals born preterm with VLBW. ²⁸ However, further investigation with larger sample sizes and experimental studies is warranted to better characterize these associations, including causal effects.

Our current understanding of how perinatal exposures may influence α‐klotho expression during adulthood is incomplete. The placenta secretes α‐klotho, and circulating fetal α‐klotho levels are elevated compared to levels in the mothers and in healthy adults. ²⁹ In term‐born healthy neonates, circulating α‐klotho levels decrease over the first few days of life. ²⁹ Studies in preeclampsia are inconsistent: α‐klotho maternal and cord blood serum levels and placental mRNA and protein expression are reduced but levels are elevated in cord blood plasma compared to normotensive pregnancies. ²⁴ , ³⁰ Preterm birth and being SGA are both associated with reduced placental α‐klotho expression and reduced neonatal circulating α‐klotho concentrations. ¹⁷ , ³¹ Thus while current evidence suggests that perinatal factors may influence expression of α‐klotho in the short term, further investigation is needed to determine whether preterm birth contributes to long‐lasting alterations to soluble α‐klotho.

Preterm birth is thought to induce a premature aging phenotype, ³² and as α‐klotho is an anti‐aging protein, suppressed α‐klotho could be an important contributor to preterm birth–related accelerated aging. Polymorphisms in the human Klotho gene are associated with an increased risk of hypertension and cardiovascular disease, but this has predominantly been described in older adults. ³³ While reductions in circulating α‐klotho levels and expression in the vasculature and heart are associated with hypertension, adverse vascular changes, and cardiac fibrosis in preclinical and observational studies, the direct correlation between BP and soluble α‐klotho in the circulation or urine has not been well described. ³⁴ , ³⁵ However, urinary levels of α‐klotho, Ang II, and Ang‐(1‐7) are generally reflective of renal expression and are independent of circulating levels. ³⁶ , ³⁷ The authors are the first to demonstrate a correlation between lower urinary soluble α‐klotho levels and higher BP in young adults born preterm with VLBW. The mechanisms behind how reduced α‐klotho contributes to hypertension are uncertain but may involve endothelial dysfunction and sodium retention. ³⁸ , ³⁹ Soluble α‐klotho concentrations, particularly in the urine, decline with worsening renal function, yet our preterm‐birth cohort had normal renal function (by estimated GFR) that was no different than that of the term‐birth cohort. ⁸

Emerging evidence suggests that the RAAS may mediate α‐klotho's effects on BP. Ang II downregulates α‐klotho expression in renal tubular cells and in endothelial cells. ¹² , ⁴⁰ Exogenous α‐klotho suppresses renin, ACE, and Ang II type 1 receptor protein expression and normalizes BP in mouse models of kidney disease. ¹¹ Thus, α‐klotho may promote sodium excretion and reduce oxidative stress by suppressing Ang II. On the other hand, Ang‐(1‐7) may protect against endothelial cell senescence by activating α‐klotho and Nrf2, but the association between α‐klotho and Ang‐(1‐7) in the kidneys is not known. ¹⁴ In addition to its inhibitory effects on Ang II, α‐klotho could upregulate Ang‐(1‐7) possibly by altering ACE or ACE2 expression or activity, though further investigation is necessary. ³⁶ Interestingly, while both the preterm‐birth and term‐birth participants demonstrated an inverse association between α‐klotho and Ang II/Ang‐(1‐7), the authors found that in the preterm‐birth cohort, this was primarily driven by lower Ang‐(1‐7)/creatinine, while in the term‐birth cohort, this was primarily driven by higher Ang II/creatinine, further supporting our theory that individuals born preterm with VLBW may have lost the potentially synergistic beneficial relationship between α‐klotho and Ang‐(1‐7) in the kidneys. Our finding of an association between SBP and urinary α‐klotho/creatinine in the preterm but not term‐born cohort further supports this idea.

Perinatal programming events have short‐ and long‐term effects on the RAAS and in particular the ACE2‐Ang‐(1‐7)‐Mas receptor pathway, though not all studies have demonstrated this consistently mostly due to methodological heterogeneity. ²⁸ , ⁴¹ Consistent with the pattern seen with α‐klotho, maternal plasma Ang‐(1‐7) levels are reduced in women with preeclampsia and in those who deliver prematurely. ⁴² , ⁴³ Preterm birth is also associated with reduced fetal circulating Ang‐(1‐7) levels. ⁴³ In our cohort, the authors previously demonstrated that preterm birth was associated with lower circulatory Ang‐(1‐7) at age 14 years. ¹⁸ In addition, lower urinary Ang‐(1‐7) levels at age 14 were associated with higher BP at age 14 and predicted higher BP in young adulthood. ¹⁸ , ²⁸ In the current analysis, the authors again showed that urinary Ang‐(1‐7) is inversely associated with BP in young adults born preterm with VLBW. This provides further evidence that Ang‐(1‐7) could in part mediate α‐klotho's effects on BP.

Our study had numerous strengths, including a large prospective preterm‐birth cohort and a term‐born peer group, comprehensive measurement of the RAAS, and robust BP measurement methodology. Limitations of our study include the small sample size in the term cohort which likely reduced the power the authors had to detect preterm‐term birth differences, limited our ability to detect effect modification by birth status, and precluded our ability to make stronger inferences about how preterm birth influences α‐klotho. The cross‐sectional analytic design and observational nature of the study preclude our ability to make causal inferences, in particular examination of the temporal and sequence of events and directionality with respect to changes in α‐klotho, Ang‐(1‐7), and BP. It is possible that preterm‐term birth differences will be detectable as the cohort ages. The authors did not assess the various isoforms of α‐klotho or its genotype. ⁴⁴ The correlation of spot urine α‐klotho concentration with 24‐hour urine content remains to be confirmed. ⁸ Future analysis of α‐klotho and RAAS components in the blood is warranted, as is measurement of fibroblast growth factor 23 and vitamin D, given their associations with α‐klotho and the RAAS. ⁴⁵ In addition, larger studies will enable us to evaluate whether sex or obesity modifies these associations. ¹⁸

In conclusion, the authors demonstrated that lower urinary soluble α‐klotho is associated with higher BP, lower urinary Ang‐(1‐7), and higher urinary Ang II/Ang‐(1‐7) in a cohort of young adults born preterm with VLBW. It is possible that suppressed renal α‐klotho combined with suppressed renal Ang‐(1‐7) could serve as a mechanism for the development of hypertension.

CONFLICT OF INTEREST

The authors have no relevant conflicts of interest to declare.

AUTHOR CONTRIBUTIONS

All authors are responsible for the reported research and agree to be accountable for all aspects of the work. All the authors have participated in the conception and design (AMS, HAS, ETJ, PAN, DID, MCC, LKW), acquisition of data (HAS, TMG, PAN, DID, MCC, LKW), or analysis and interpretation of data (all authors), and drafting or revising the manuscript (all authors), and approved the submitted manuscript (all authors).

South AM, Shaltout HA, Gwathmey TM, et al. Lower urinary α‐Klotho is associated with lower angiotensin‐(1‐7) and higher blood pressure in young adults born preterm with very low birthweight. J Clin Hypertens. 2020;22:1033–1040. 10.1111/jch.13897

DATA AVAILABILITY STATEMENT

Data from this study are available upon request.