Regulation of Vascular Tone of Preglomerular Renal Vasculature by Caldesmon

Abstract

Background

Caldesmon is an actomyosin‐tropomyosin–binding protein with pleiotropic functions that generally acts as an inhibitor of actomyosin ATPase activity. This study aims to explore the role of caldesmon in regulation of preglomerular vascular tone.

Methods

Expression of caldesmon, myosin light chain, and α‐smooth muscle actin was assessed by western blot of interlobar arteries (IAs) from young and old mice or human renal arteries from patients who underwent nephrectomy or tumor enucleation. IAs were obtained from wild‐type or heterozygous mice, carrying a mutant allele that ablates both caldesmon isoforms (caldesmon ^+/−). In IAs, the vasoreactivity to α₁‐agonist phenylephrine, the soluble guanylate cyclase activator cinaciguat, and inhibitors of RhoA‐associated protein kinase Y27632 or p21‐activated protein kinase 1 IPA‐3 were tested by wire myography.

Results

Caldesmon was expressed in all preglomerular vessels of mice and humans. In caldesmon ^+/− IAs of young mice, caldesmon protein was 50% less than in IAs of wild‐type mice. In IAs of old mice, caldesmon was less than in IAs of young mice and not significantly different between caldesmon ^+/− and wild‐type. Caldesmon mutation did not affect phenylephrine‐induced contraction or cinaciguat sensitivity but induced a rightward shift of Y27632‐induced relaxation in IAs of young and old mice and a reduction in IPA‐3–induced relaxation by 50% only in IAs of young mice. In human renal arteries, caldesmon expression negatively correlated with age and the presence of renal inflammation. In the inflammation group, a reduction in estimated glomerular filtration rate was detected.

Conclusions

This study provides cumulative evidence that caldesmon is a marker of age and inflammation‐related changes in preglomerular vascular function. Our work supports the view that in the renal vasculature, caldesmon expression is age dependent and determines its sensitivity to the RhoA‐associated protein kinase/p21‐activated protein kinase 1 regulatory pathway.

Nonstandard Abbreviations and Acronyms

ERK

extracellular signal‐regulated kinase

IA

interlobar artery

K⁺ PSS

high‐potassium physiological salt solution

l‐NAME

Nω‐nitro‐l‐arginine methylester hydrochloride

MLC₂₀

20‐kDa myosin regulatory light chain

NO

nitric oxide

PAK1

p21‐activated protein kinase 1

PSS

physiological salt solution

ROCK

RhoA‐associated protein kinase

WT

wild‐type

Research Perspective.

What Is New?

• Our study demonstrated that, similar to murine preglomerular vasculature, human intrarenal arteries express the protein caldesmon.

• Using a mutant caldesmon mouse line, we demonstrated that caldesmon mutation counteracts the RhoA‐associated protein kinase/p21‐activated protein kinase 1–dependent contractile pathway in murine interlobar arteries.

What Question Should Be Addressed Next?

• Future studies with larger patient cohorts would allow detailed analyses aiming to uncover the clinical conditions responsible for the reduction of the caldesmon protein with advanced age and inflammatory diseases, such as pyonephrosis or pyelonephritis. These investigations will highlight the molecular mechanisms of renal dysfunction caused by imbalanced reactivity of preglomerular vasculature and may provide a foundation for developing therapeutic strategies for managing renal vascular pathology.

Acute and chronic renal dysfunction is associated with an increased risk for morbidity and progression to end‐stage renal failure with a need for renal replacement therapy and is associated with an increased mortality rate. ¹ Renal dysfunction encompasses a series of intrarenal and systemic events, including diminished renal perfusion; proteinuria; changes in water, electrolyte, and acid–base balance; and dysregulation of the renin–angiotensin–aldosterone axis and sympathetic nervous system. ² Meanwhile, it is accepted that one of the most important elements responsible for the occurrence of renal pathology relates to the maladaptive changes in the preglomerular vessels, which play a primary role in the control of renal vascular resistance. ³

The vascular tone of the preglomerular renal vessels is primarily regulated by the reversible phosphorylation of the 20‐kDa regulatory light chains (MLC₂₀) of smooth muscle myosin thick filaments. ⁴ , ⁵ , ⁶ Phosphorylation occurs in response to elevated intracellular Ca²⁺, which leads to the activation of specific Ca²⁺‐calmodulin–dependent myosin light‐chain kinase and phosphorylation of MLC₂₀ at serine19. ⁴ This increases the ATPase activity of the actomyosin complex, leading to cross‐bridge cycling and force generation. MLC₂₀ phosphorylation might also be catalyzed in a myosin light‐chain kinase–independent manner via phosphorylation and activation of the p90 ribosomal S6 kinase, ⁷ a kinase recently shown to be highly phosphorylated in human renal vasculature. ⁴ Other kinases, such as zipper‐interacting protein kinase, integrin‐linked kinase, and RhoA‐associated protein kinase (ROCK), have also been reported to phosphorylate MLC₂₀ in smooth muscle cells directly ⁸ , ⁹ and are proposed to be involved in vascular tone generation. ¹⁰ , ¹¹ Furthermore, to date, it remains elusive whether other mechanisms, such as interference of the MLC₂₀ phosphorylation by the small molecular kinase–related protein telokin, play a role in the regulation of the renal vasculature tone. ¹²

In addition, the smooth muscle actomyosin ATPase activity is regulated by the actin–myosin–tropomyosin–calmodulin–binding protein, caldesmon. It is expressed in 2 isoforms, the smooth muscle–specific high‐molecular‐weight isoform and the low‐molecular‐weight isoform, which is expressed in many nonmuscle cells. ¹³ The isoforms are generated by alternative splicing. ¹³ Caldesmon was isolated from smooth muscle >4 decades ago. ¹⁴ In the absence of Ca²⁺, it inhibits smooth muscle actomyosin ATPase, even of fully phosphorylated myosin. ¹⁵ Based on this, caldesmon was proposed to act as a molecular brake of smooth muscle contraction. ¹³ Ablation of caldesmon by gene targeting in fetal bladder and aorta resulted in an increased basal tone but unexpectedly in a reduced force‐bearing capacity. ¹⁶ The same observation was made in adult carotid arteries using small interfering RNA. ¹⁷ The latter was ascribed to loss of caldesmon’s putative structural role as it tethers actin and myosin filaments of the contractile domain, facilitating cross‐bridge formation and force generation (reviewed in G CJaP ¹⁸ ). The binding of Ca²⁺‐calmodulin to the C‐terminal domain reverses its inhibitory effect on actomyosin ATPase activity (reviewed in Kim et al ¹⁹ ). Moreover, numerous publications demonstrated that caldesmon can be phosphorylated by a number of protein kinases, including extracellular signal‐regulated kinase (ERK) 1/2, p21‐activated protein kinase 3, and calcium/calmodulin‐dependent protein kinase II. ²⁰ , ²¹ , ²² , ²³ These phosphorylations are proposed to alleviate the inhibitory function of the protein in muscle and nonmuscle cells (reviewed in Wang et al ²⁴ ).

Despite extensive investigation, the role of caldesmon in modulating vascular tone within the renal circulation, particularly in the preglomerular vasculature, remains incompletely understood. In our previous work, we observed that the mutation of caldesmon in heterozygous mice induces an early onset of stretch‐induced tone in basilar arteries from old animals (aged 18–20 months), which was ROCK‐sensitive and observed in wild‐type arteries in senescent animals aged >23 months. ²⁵ This prompted us to extend these investigations to the vasculature of human and murine renal preglomerular renal arteries, as their tone to a large extent depends on ROCK. ⁴ , ²⁵ , ²⁶ In this study, we used a gene targeting approach for global low‐molecular‐weight and high‐molecular‐weight caldesmon ablation ¹⁶ and conducted a comparative analysis of caldesmon expression and vascular tone in interlobar arteries (IAs) from heterozygotes and wild‐type mice. Additionally, we examined age‐related expression patterns of caldesmon in intrarenal arteries from patients who had undergone nephrectomy or enucleation due to carcinoma or inflammation. These expression patterns were compared with estimated glomerular filtration rate (eGFR) values. We also investigated caldesmon localization within the renal vasculature, including the capillary network and tubular structures, using kidney samples from the same patient cohort. We hypothesize that caldesmon expression serves as a critical determinant of preglomerular vascular tone and plays a pivotal role in the maintenance of normal renal function.

Methods

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics Statement

All procedures involving patient tissue were approved by the Ethics Committee of the Medical School Brandenburg, Germany (Reference No. E‐01‐20 200 921) and the Medical University Varna, Bulgaria (Reference No. 140/01.02.2024). They conformed to the principles outlined in the Declaration of Helsinki. Human resected specimens were obtained from patients who underwent nephrectomy because of a renal or pelvic tumor or renal inflammation (pyonephrosis, pyelonephritis, or ureteral inflammation) or tumor enucleation after informed consent (detailed description in Table S1). Tissue samples from 48 patients (21 men, 27 women; age, 65.0±12.5 years) were investigated. The mean age of men was 66.5±12.2 years and of women 63.9±12.0 years, respectively (Table S1).

All animal experimental mouse protocols were approved by the State Office for Nature, Environment and Customer Protection, North Rhine‐Westphalia (AZ 84–02.05.50.15.029). All experiments were performed following the European Community guidelines (Directive 2010/63/EU) and the Animal Research: Reporting of In Vivo Experiments guidelines. Young (aged 12–23 weeks) and old (aged 22–24 months) wild‐type (WT) and heterozygous mice carrying a global mutation for caldesmon (caldesmon^+/−) of either sex, all with C57BL/6N background, were kept at a 12‐hour dark/light cycle at the Center of Molecular Medicine, Cologne, Germany. The mutant caldesmon^+/− mice were generated as previously described. ¹⁶ Because of prenatal lethality of homozygous caldesmon^−/− mice, ¹⁶ all experiments were performed in arteries of heterozygotes caldesmon^+/− versus homozygotes caldesmon^+/+ WT mice. All animals were euthanized by decapitation under isoflurane anesthesia.

Detailed description of methods, such as wire myography, protein extraction, immunoblotting, and immunohistochemistry, as well as solutions, buffers, and primary and secondary antibodies used in this work, is available in Data S1, Tables S2 and S3, and Figure S3.

Assessment of eGFR

eGFR values were assessed in patients with carcinoma/malignant diseases (n=24) versus patients with inflammation (n=11; 1 participant excluded) by the creatinine‐based Chronic Kidney Disease Epidemiology Collaboration equation (https://www.kidney.org/ckd‐epi‐creatinine‐equation‐2021) ²⁷ ;

$$\begin{array}{c}\text{eGFRcr}=142\times \min \left(\mathrm{Scr}/\kappa ,1\right)\alpha \times \max \left(\mathrm{Scr}/\kappa ,1\right)\hfill \\ –1.200\times {0.9938}^{\mathrm{Age}}\times 1.012\left[\text{if female}\mathrm{sex}\right]\hfill \end{array}$$

Where:

Scr=standardized serum creatinine in mg/dL.

κ=0.7 (females) or 0.9 (males).

α=−0.241 (female) or −0.302 (male).

min (Scr/κ, 1) is the minimum of serum creatinine/κ or 1.0.

max (Scr/κ, 1) is the maximum of serum creatinine/κ or 1.0.

Age (years).

Statistical Analysis

Results are expressed as mean±SEM. Phenylephrine contraction is expressed as percentage of force relative to maximal precontraction achieved after stimulation of the vessels with 60 mM K⁺‐PSS and accepted as 100%. Unless otherwise indicated, n is the number of animals. The significance level was set at P<0.05 and was tested by Student’s t test when comparing 2 pairs or by 2‐way ANOVA for multiple comparisons, followed by Tukey’s multiple comparison posttest. Correlative analysis and linear regression were analyzed using Spearman’s correlation coefficient, r, and values P<0.05 were considered to indicate statistical significance. All statistical calculations were performed with Prism software (GraphPad Software, La Jolla, CA).

Correlative Analysis of Caldesmon, α‐Actin and MLC₂₀ — Expression

The observer‐blind correlative analysis of caldesmon, α‐actin, and MLC₂₀ expression was performed in a group of 37 patients. During this analysis, the only known information about the patients was their age, plotted versus the values of caldesmon/GAPDH, α‐actin/GAPDH, or MLC₂₀/GAPDH ratio. In the samples obtained from patients with inflammation (pyonephrosis and pyelonephritis), the immunoreactive signal was weak or even below the detection limit of the western blot analysis. Samples with no detectable caldesmon signal were excluded from the final calculation. Therefore, the number of participants in this analysis is 30.

In a separate analysis, the levels of caldesmon/GAPDH ratio in patients with carcinoma/malignant diseases (n=24) were compared versus the same parameter in patients with inflammation (n=12) using the Student–Fisher unpaired t test. One patient with pyonephrosis was excluded from the calculation because the caldesmon/GAPDH ratio was >2‐fold higher than the SD.

Results

Localization of the Caldesmon Protein in Murine Renal Preglomerular Vasculature

In kidney sections from young and old WT and caldesmon^+/− animals, immunoreactive anti‐caldesmon signals were observed along the complete preglomerular vasculature, that is, interlobar, arcuate, and interlobular arteries (Figure 1A). Caldesmon signal intensity in young WT arteries appeared stronger than in arteries from caldesmon^+/− littermates (Figure 1B) with a comparable diameter. At advanced age, caldesmon immunoreactive signals were reduced in both WT and caldesmon^+/− IAs to a comparable degree (Figure 1C). In young and old WT and caldesmon^+/− animals, we observed caldesmon expression in the vascular endothelium. Consistent with our results obtained in preglomerular large kidney vessels, we found a similar reduction of caldesmon protein with advanced age or caldesmon mutation in renal resistance arterioles (Figure S1).

Figure 1. Caldesmon localization in murine renal preglomerular arteries and arterioles.

A, Caldesmon localization along the kidney preglomerular vascular tree. B, C, Localization of caldesmon in young and old interlobar arteries from WT (B [i] and C [i]) and caldesmon^+/− (B [ii] and C [ii]) animals. Caldesmon immunoreactive signals were detected by incubation with anti‐caldesmon antibody (gray) and those from endothelial cells by incubation with anti‐CD31 (green). Signals in blue represent nuclei stained by DAPI. Original confocal images from experiments (n=3 [3 replicates each]). CaD indicates caldesmon; and WT, wild‐type.

Caldesmon Expression in Murine IAs From WT and Caldesmon ^+/− Mice

Consistent with the immunohistochemical experiments, we observed that caldesmon immunoreactive signals were ≈50% less in the homogenates of interlobar arteries from young caldesmon^+/− animals compared with WT littermates in western blots (Figure 2A and 2C). Aging reduced caldesmon immunoreactivity in both genotypes to a degree that no differences between WT and caldesmon^+/− mice were observed (Figure 2A and 2C). No difference in expression pattern of α‐smooth muscle actin, MLC₂₀ (used as a marker for expression of the myosin filament), and the targeting subunit of the myosin light‐chain phosphatase MYPT1, used as a marker for expression of the enzyme myosin light‐chain phosphatase, were detected between IAs of WT and caldesmon^+/− animals (Figure 2A, 2B, 2D through 2F).

Figure 2. Expression of caldesmon in interlobar arteries from wild‐type and heterozygote animals carrying a global low‐molecular‐weight and high‐molecular‐weight caldesmon mutation (caldesmon^+/−).

A–D, Original western blots and statistical evaluation representing the expression levels of caldesmon in young (A, C; n=9–11) and old (B, C; n=7–9) interlobar arteries isolated from WT and caldesmon^+/− animals. D–F, Expression of α‐actin, MLC₂₀, and targeting subunit of myosin light‐chain phosphatase, MYPT1, in the same vessels. Expression levels were determined as the ratio of immunoreactive signal of gene of interest toward the signal of GAPDH used as a housekeeping protein and normalized to the mean ratio in young WT arteries. Data presented as mean±SEM; dots indicate individual animals (arteries) used in these experiments. P values calculated using 2‐way ANOVA. CaD indicates caldesmon; MLC_2. 20‐kDa myosin light chain; MYPT1, myosin phosphatase‐I targeting subunit; and WT, wild‐type.

Vascular Diameter and Stretch‐Induced Tone of Murine IAs From WT and Caldesmon ^+/− Animals

The vascular lumen diameter of murine IAs, determined at a wall tension corresponding to a transmural pressure of IC100 (see Methods) was 296±7 μm in young IAs from WT versus 262±5 μm in vessels from caldesmon^+/− animals (P<0.01; n=23–26 vessels from 12 to 14 animals; 2‐way ANOVA). The lumen diameter of old WT vessels (297±7 μm) was similar to that of young IAs, whereas that of old IAs from caldesmon^+/− was significantly higher than in young caldesmon^+/− IAs (317±8 μm; P<0.001). IA lumen diameters did not differ significantly between old WT and caldesmon^+/− animals (P=0.09; n=14–19 vessels from 8 to 10 animals; Figure 3A and 3B).

Figure 3. Passive force and NO‐dependent endothelial reactivity of interlobar arteries from WT and caldesmon^+/− animals.

A, B, Length–tension relationships (A) and stretch‐induced tone (B) of young and old interlobar arteries from WT and caldesmon^+/− animals. Summary from experiments on 6‐11 young and 6‐11 old WT and caldesmon^+/− interlobar arteries. C, Time course of the passive force 20 minutes after normalization (see Methods). (D) NO‐dependent tone of interlobar arteries from WT young (i), WT old (ii), caldesmon^+/− young (iii), and caldesmon^+/−old (iv) animals. Dots represent a number of experiments (animals). P values calculated using 2‐way ANOVA (B, C) or paired t‐test (D). CaD indicates caldesmon; l‐NAME, Nω‐nitro‐l‐arginine methylester hydrochloride; and WT, wild‐type.

In young IAs from WT mice, stretch‐induced tone at IC90 (see Methods) was 2.81±0.4 mN and amounted to 3.22±0.3 mN after 20 minutes of equilibration. The values of stretch‐induced tone did not differ in young caldesmon^+/− IAs and were respectively 2.79±0.1 mN at IC90 and 3.27±1.3 mN after 20 minutes of equilibration (P>0.05; 2‐way ANOVA; Figure 3C). In old IAs from WT mice, these parameters were 3.28±0.8 mN at IC90 and 3.33±0.5 mN after 20 minutes of equilibration (P>0.05 young versus old; 2‐way ANOVA). By contrast, in IAs from old caldesmon^+/− mice, the tone measured was 3.98±0.4 mN and 4.38±0.4 mN after the same equilibration time (P<0.05; old caldesmon^+/− versus young caldesmon^+/−; 2‐way ANOVA; Figure 3C).

Effect of Nitric Oxide Inhibition on Stretch‐Induced Tone in Young and Old WT and Caldesmon ^+/− Interlobar Arteries

We also determined stretch‐induced tone in all groups after inhibition of endogenous nitric oxide (NO) release. IAs from young WT or caldesmon^+/− animals were incubated for 20 minutes with the pan‐NO synthase inhibitor Nω‐nitro‐l‐arginine methylester hydrochloride (l‐NAME) after the first stimulation with K60. As shown previously, ²⁸ , ²⁹ NO synthase inhibition had no significant effect on tone in young IAs (Figure 3D). In contrast, in old WT vessels, NO inhibition significantly increased tone in 6 of 8 experiments (P>0.01; paired t test; Figure 3D [ii]). In only 3 of 10 experiments, l‐NAME treatment increased tone in IAs from old caldesmon^+/− animals (P>0.05; paired t test; Figure 3D [iv]).

Responsiveness to α₁‐Receptor Activation in IAs From WT and Caldesmon ^+/− Mice

Challenging IAs from young and old imice IAs with cumulatively increasing concentrations of the α₁‐receptor agonist phenylephrine induced a concentration‐dependent increase in force (Figure 4A, 4B, 4D, and 4E). Inhibition of endogenous NO release by l‐NAME induced a small leftward shift in the phenylephrine concentration–response relation, which did not reach the level of statistical significance (Figure 4C, F).

Figure 4. Reactivity to α₁‐agonist of young and old interlobar arteries from WT and caldesmon^+/− animals.

A–E, Original force tracings representing the contractile effect of α₁₋agonist phenylephrine in interlobar arteries from young (A, B) and old (D, E) WT and caldesmon^+/− animals. Statistical summary of the effect of cumulative application of phenylephrine under control conditions and in the presence of 100 μmoL/L l‐NAME in young (C) and old (F) WT caldesmon^+/− interlobar arteries. (C) Summary of results from n=8 WT and n=11 CaD^+/− young interlobar arteries. (F) Summary of results from n=6 WT and n=4 old WT or CaD^+/− old interlobar arteries. Data represented as percentage from the force achieved after stimulation of the vessels with 60 mM K⁺‐PSS and accepted as 100%. Data in panels (C) and (F) were analyzed by 2‐way ANOVA testing for the effects of the factors genotype and l‐NAME treatment on pD₂ or F _max for phenylephrine. Statistics: phenylephrine F _max young: P=0.7, caldesmon^+/−_vs WT; P=0.3, caldesmon^+/−_(l‐NAME) vs WT (l‐NAME); P=0.94, caldesmon^+/− vs caldesmon^+/−_(l‐NAME); P=0.58, WT vs WT (l‐NAME). Phenylephrine F _max old: P=0.45, caldesmon^+/−_vs WT; P=0.75, caldesmon^+/−_(l‐NAME) vs WT (l‐NAME); P=0.85, caldesmon^+/− vs caldesmon^+/−_(l‐NAME); P=0.99, WT vs WT (l‐NAME). Phenylephrine pD₂ young: P=0.77, caldesmon^+/−_vs WT; P=0.51, caldesmon^+/−_(l‐NAME) vs WT (l‐NAME); P=0.02, caldesmon^+/− vs caldesmon^+/−_(l‐NAME); P=0.35, WT vs WT (l‐NAME). Phenylephrine pD₂ old: P=0.97, caldesmon^+/−_vs WT; P=0.92, caldesmon^+/−_(l‐NAME) vs WT (l‐NAME); P=0.49, caldesmon^+/− vs caldesmon^+/−_(l‐NAME); P=0.22, WT vs WT (l‐NAME). CaD caldesmon; F _max, maximal relaxation (%); K⁺‐PSS, l‐NAME, Nω‐nitro‐l‐arginine methylester hydrochloride; and WT, wild‐type.

Endothelial Responsiveness to Acetylcholine in Young and Old IAs From WT and Caldesmon ^+/− Mice

Endothelial responsiveness of submaximally activated IAs was tested using acetylcholine. After phenylephrine stimulation, preparations from both age and genotype groups were treated by application of cumulatively increasing concentrations of acetylcholine. In all age and genotype groups, acetylcholine induced a concentration‐dependent relaxation consisting of an initial fast and a slow component (Figures 5A through 5D and 6A through 6D). No difference in acetylcholine‐induced relaxation was observed between the genotypes (Figures 5E, 5F, 6E, and 6F; Table 1).

Figure 5. Reactivity to acetylcholine of young interlobar arteries from WT and caldesmon^+/− animals.

A–D, Original force tracings representing the effect of acetylcholine under control conditions (A, C) and in the presence of 100 μmoL/L l‐NAME (B, D) in interlobar arteries from young WT (A, B) and caldesmon^+/− (C, D) animals. Vessels preconsticted submaximally by 0.3 μmoL/L and at steady state treated by cumulative application of acetylcholine for 5 to 8 minutes each concentration. (E, F). Statistical summary of the relaxant effect of acetylcholine. Observed effect showed a biphasic pattern, and therefore (E) represents a summary of the fast and (F) he slow steady‐state component of the observed relaxation. Data in young (E) Summary of results from n=8 WT and n=7 caldesmon^+/− young interlobar arteries. (F) Summary of results from n=12 WT and n=13 old WT or caldesmon^+/− old interlobar arteries. represented as a percentage from the steady‐state force reached after administration of 0.3 μmoL/L phenylephrine and accepted as 100%. Data in panels (E) and (F) were analyzed by 2‐way ANOVA testing for the effects of the factors genotype and l‐NAME treatment on pD₂ or F _max for the fast and slow relaxation induced by acetylcholine in young interlobar arteries. Statistics: F _max fast relaxation: P=0.98, caldesmon^+/−_young vs WT young; P=0.3, caldesmon^+/−_young (l‐NAME) vs WT_young (l‐NAME); P=0.007, caldesmon^+/−_young vs caldesmon^+/−_young (l‐NAME); P=0.21, WT_young vs WT_young (l‐NAME). F _max slow relaxation: P=0.96, caldesmon^+/−_young vs WT young; P=0.74, caldesmon^+/−_young (l‐NAME) vs WT_young (l‐NAME); P=0.0006, caldesmon^+/−_young vs caldesmon^+/−_young (l‐NAME); P=0.01, WT_young vs WT_young (l‐NAME). pD₂ fast relaxation: P=0.67, caldesmon^+/−_young vs WT young; P=0.48, caldesmon^+/−_young (l‐NAME) vs WT_young (l‐NAME); P=0.42, caldesmon^+/−_young vs caldesmon^+/−_young (l‐NAME); P=0.65, WT_young vs WT_young (l‐NAME). pD₂ slow relaxation: P=0.52, caldesmon^+/−_young vs WT young; P=0.06, caldesmon^+/−_young (l‐NAME) vs WT_young (l‐NAME); P=0.16, caldesmon^+/−_young vs caldesmon^+/−_young (l‐NAME); P=0.03, WT young vs WT_young (l‐NAME). Caldesmon indicates caldesmon; F _max, maximal relaxation (%); l‐NAME, Nω‐nitro‐l‐arginine methylester hydrochloride; and WT, wild‐type.

Figure 6. Reactivity acetylcholine of old interlobar arteries from WT and caldesmon^+/− animals.

A–D, Original force tracings representing the cumulative effect of acetylcholine under control conditions (A, C) and in the presence of 100 μmoL/L l‐NAME (B, D) in interlobar arteries from old WT and caldesmon^+/− animals. Vessels were preconsticted submaximally by 0.3 μmoL/L phenylephrine and the relaxant effect of cumulative application of acetylcholine for 5 to 8 minutes each concentration. (E, F) Statistic summary of the fast and (F) the slow steady‐state component of the observed relaxation. Data in young and old represented as a percentage from the steady‐state force reached after administration of 0.3 μmoL/L phenylephrine and accepted as 100%. Data in panels (E) and (F) were analyzed by 2‐way ANOVA testing for the effects of the factors genotype and l‐NAME treatment on pD₂ or F _max for the fast and slow relaxation induced by acetylcholine in old interlobar arteries. Statistics: F _max fast relaxation: P=0.53, caldesmon^+/−_old vs WT old; P=0.53, caldesmon^+/−_old (l‐NAME) vs WT_old (l‐NAME); P=0.005, caldesmon^+/−_old vs caldesmon+/−_old (L‐NAME); P=0.007, WT_old vs. WT_old (L‐NAME). F _max Slow relaxation:: P=0.049, caldesmon^+/−_old vs WT old; P=0.01, caldesmon^+/−_old (l‐NAME) vs WT_old (l‐NAME); P<0.0001, caldesmon^+/−_old vs caldesmon^+/−_old (l‐NAME); P<0.0001, WT_old vs WT_old (l‐NAME). pD₂ fast relaxation:: P=0.86, caldesmon^+/−_old vs WT old; P=0.38, caldesmon^+/−_old (l‐NAME) vs WT_old (l‐NAME); P=0.53, caldesmon^+/−_old vs caldesmon^+/−_old (l‐NAME); P=0.68, WT_old vs WT_old (l‐NAME). pD₂ slow relaxation:: P=0.63, caldesmon^+/−_old vs WT old; P=0.11, caldesmon^+/−_old (l‐NAME) vs WT_old (l‐NAME); P=0.95, caldesmon^+/−_old vs caldesmon+/−_old (l‐NAME); P=0.03, WT_old vs WT_old (l‐NAME). Caldesmon indicates caldesmon; F _max, maximal relaxation (%); l‐NAME, Nω‐nitro‐l‐arginine methylester hydrochloride; PSS, physiological salt solution; and WT, wild‐type.

Table 1.

Acetylcholine‐Induced Relaxation in Murine Interlobar Arteries From Young and Old WT and Caldesmon^+/− Animals

	pD2_fast	F max_fast	pD2_slow	F max_slow	No.
Controls, WT_young,	−7.4±0.3	−84.0±5.2	−7.0±0.4	−82.7±6.0	8
l‐NAME, WT_young	−7.6±0.8	−73.0±7.1	−6.4±0.3	−53.0±13.0	7
Controls, caldesmon+/−_young	−7.6±0.8	−84.0±3.4	−7.1±0.4	−81.9±4.0	13
l‐NAME Caldesmon+/−_young	−7.4±0.5.	−64.9±6.6	−6.8±0.7	−50.5±6.9	12
Controls WT_old	−7.3±0.1	−83.6±6.4	−7.0±0.2	−63.7±11.3	8
l‐NAME WT_old	−7.4±0.35	−56.9±6.8	−6.8±0.2	−7.9±4.0	10
Controls Caldesmon+/−_old	−7.4±0.2	−89.3±3.5	−6.8±0.1	−85.1±4.0	10
l‐NAME Caldesmon+/−_old	−7.2±0.25	−62.1±6.9	−6.7±0.2	−34.7±8.5	10

F _max indicates maximal relaxation in %; l‐NAME, Nω‐nitro‐L‐arginine‐methylester‐hydrochlorid; pD₂, negative logarithm of the concentration of the compound able to induce a 50%‐relaxation; and WT, wild‐type.

In young IAs, the fast component of the relaxation was not sensitive to NO synthase inhibition by 100 μmoL/L l‐NAME (Figure 5E; Table 1), whereas the slow component was reduced in both genotypes by ≈20% (Figure 5F; Table 1). In IAs from old WT mice, l‐NAME reduced the fast relaxation component by ≈20% (Figure 6E) and nearly completely abolished the slow component (Figure 6F). Interestingly, in IAs from old caldesmon^+/− mice, the slow relaxation was inhibited by only 50% (Figure 6F; Table 1).

IAs From Young and Old Animals Relaxed to Direct Soluble Guanylate Cyclase Activation and Inhibition of ROCK and p21‐Activated Protein Kinase 1

We next tested the reactivity of young and old IAs toward pharmacological activation of soluble guanylate cyclase and inhibition of the protein kinases ROCK and p21‐activated protein kinase 1 (PAK1). In these experiments, submaximally stimulated IAs (phenylephrine 0.3 μmoL/L) of both ages and genotypes were treated by cumulatively increasing concentrations of the direct activator of the soluble guanylate cyclase cinaciguat (concentration range, 10⁻¹¹–10⁻⁵ moL/L), the RhoA‐kinase inhibitor Y27632 (concentration range, 10⁻⁸–10⁻⁵ moL/L), or the PAK1‐inhibitor IPA‐3 (concentration range, 10⁻⁸–10⁻⁴ moL/L). All compounds induced a concentration‐dependent relaxation. IAs from all animals showed the same degree of relaxation when treated with cinaciguat (Figure 7; Table 2). The pD₂ value or the relaxation induced by the ROCK inhibitor Y27632 was higher in IAs from WT than from caldesmon^+/− mice of both ages, indicating a lower potency (Figure 8; Table 2). Furthermore, in young caldesmon^+/− IAs, the efficacy of PAK1 inhibition by 3‐indolepropionic acid was significantly attenuated (Figure 9A through 9C; Table 2). Interestingly, IPA‐3 relaxed old IAs without a statistically significant difference between genotypes (Figure 9D through 9F; Table 2).

Figure 7. Reactivity of interlobar arteries to a direct soluble guanylate cyclase activation.

A, B, D, E, Original force tracings and statistical evaluation (C, F) of the relaxant effect of the direct activator of the soluble guanylate cyclase, cinaciguat (n=7–12 young and 9–10 old WT and caldesmon^+/− interlobar arteries). Data represented as a percentage from the steady‐state force reached after application of 0.3 μmoL/L phenylephrine and accepted as 100%. Data in panels (C) and (F) were analyzed by 2‐way ANOVA testing for the effects of the genotype on pD₂ or F _max for the cinaciguat‐induced relaxation in young (C) and old (F) interlobar arteries. Statistics: F _max cinacigat relaxation: P=0.99, caldesmon^+/−_young vs WT young; P=0.26, caldesmon^+/−_old vs WT_old; P=0.8, caldesmon^+/−_young vs caldesmon^+/−_old; P=0.39, WT_young vs WT old. pD₂ cinaciguat relaxation:: P=0.24, caldesmon^+/−_yound vs WT young; P=0.2, caldesmon^+/−_old vs WT_old; P=0.02, caldesmon^+/−_young vs caldesmon^+/−_old; P=0.05, WT_young vs WT old. CaD indicates caldesmon; F _max, maximal relaxation (%); n.s., not significant; PSS, physiological salt solution; and WT, wild‐type animals.

Table 2.

Cinaciguat‐, Y27632‐ and IPA‐3–Induced Relaxation in Murine Interlobar Arteries From Young and Old WT and Caldesmon^+/− Animals

	pD2 (young)	F max (young)	n	pD2 (old)	F max (old)	No.
Cinaciguat_WT	−8.1±0.3	−75.9±5.6	7	−7.5±0.2	−81.3±2.9	9
Cinaciguat_Caldesmon+/−	−8.4±0.1	−76.1±4.1	12	−7.9±0.2§	−75.0±3.6	10
Y27632_WT	−6.3±0.2	−84.4±3.5	5	−6.1±0.2	−80.1±4.1	11
Y27632_Caldesmon+/−	−5.8±0.1	−84.9±3.9	9	−5.7±0.1	−85.4±1.8	12
IPA‐2_WT	−5.3±0.2	−83.1±5.1	6	−5.2±0.5	−73.7±9.3	3
IPA‐3_Caldesmon+/−	−5.1±0.3	−44.4±6.3	5	−5.0±0.3	−67.4±8.0	4

F _max, maximal relaxation (%); IPA‐3, pD₂, negative logarithm of the concentration of the compound able to induce a 50% relaxation; and WT, wild‐type.

Figure 8. Reactivity of interlobar arteries to ROCK inhibition.

A, B, D, E, Original force tracings and statistical evaluation (C, F) of the relaxant effect of the ROCK inhibitor, Y27632 in young (A, B; black and blue) and old (D, E; gray and sky‐blue) interlobar arteries from WT and caldesmon^+/− animals (n=6–12 young and 9–10 old WT and caldesmon^+/− interlobar arteries). Data represented as % from the steady state force reached after application of 0.3 μmoL/L Phenylephrine and accepted as 100%. Data in panels (C) and (F) were analyzed by two‐way ANOVA testing for the effects of the genotype on pD₂ or F _max for the Y27632‐induced relaxation in young (C) and old (F) interlobar arteries. Statistics: F _max Y27632 relaxation: P=0.48, caldesmon^+/−_young vs WT young; P=0.92, caldesmon^+/− old vs WT_old; P=0.8, caldesmon^+/−_young vs caldesmon^+/− old; P=0.39, WT_young vs WT old. pD₂ Y27632 relaxation: P=0.002, caldesmon^+/−_young vs WT young; P=0.006, caldesmon^+/−_old vs WT old; P=0.45, caldesmon^+/−_young vs caldesmon^+/−_old; P=0.14, WT_young vs WT old. CaD indicates caldesmon; F _max, maximal relaxation (%); PAK1, p21‐activated protein kinase 1; ROCK, RhoA‐associated protein kinase; and WT, wild type animals.

Figure 9. Reactivity of interlobar arteries to pharmacological inhibition of PAK1.

(A, B, D, E) Original force tracings and statistical evaluation (E, F) of the relaxant effect of the specific PAK1‐inhibitor IPA‐3 on young (black and blue) and old (gray and sky‐blue) in interlobar arteries from WT and caldesmon^+/− animals (n=5–4 young and 6–7 old WT and caldesmon^+/− interlobar arteries). Data represented as a percentage from the steady‐state force reached after application of 0.3 μmoL/L Phenylephrine and accepted as 100%. Data in panels (C) and (F) were analyzed by 2‐way ANOVA testing for the effects of the genotype on pD₂ or F _max for the IPA‐3–induced relaxation in young (C) and old (F) interlobar arteries. Statistics: F _max IPA‐3 relaxation: P=0.0005, caldesmon^+/−_yound vs WT young; P=0.57, caldesmon^+/−_old vs WT_old; P=0.03, caldesmon^+/−_young vs caldesmon^+/−_old; P=0.37, WT_young vs WT old. pD₂ IPA‐3 relaxation: P=0.82, caldesmon^+/−_young vs WT young; P=0.16, caldesmon^+/−_old vs WT_old; P=0.1, caldesmon^+/−_young vs caldesmon^+/−_old; P=0.79, WT_young vs WT old. Caldesmon indicates caldesmon; IPA‐3, 3‐indolepropionic acid; PAK1, p21‐activated protein kinase 1; and WT, wild type animals.

Caldesmon Localization in Human Kidney Cortical and Medullar Vasculature

As in murine cortical kidney sections (see Figure 1), caldesmon immunoreactivity was predominantly observed in smooth muscle layers and vascular endothelium of human cortical kidney sections (Figure 10A). Significant caldesmon immunoreactive signals were also observed in the glomerular endothelia and Bowman’s capsule (Figure 10B). In the medulla, caldesmon localized predominantly to vascular smooth muscle layers and less to the endothelium of vasa recta but not to the endothelia of peritubular capillaries (Figure S2).

Figure 10. Caldesmon is present in smooth muscle and endothelial cells of the human intrarenal arteries.

A, Representative confocal image of a human kidney section showing the wall of an arcuate artery. B, Section depicting an interlobular artery connected to an afferent arteriole. Smooth muscle cells were visualized using an anti–α‐smooth muscle Actin antibody (red), endothelial cells using an anti‐CD31 antibody (green), and caldesmon with a pseudocolored gray channel. Nuclei were counterstained with DAPI (blue). CaD, Caldesmon; α‐SMA, α‐smooth muscle Actin; and WT, Wild type animals.

Caldesmon Expression in Human Renal Vasculature Declines in Advanced Age and Inflammation

Expression levels of caldesmon were assessed in homogenates from human preglomerular vessels using western blot analysis. With advanced age, we found a negative correlation of the caldesmon/GAPDH ratio (n=30; P=0.01; Figure 11A and 11B). We also found a trend for negative correlation with age of the α‐actin/GAPDH ratio (n=30; P=0.2; Figure 11A and 11C) and a positive correlation between age and MLC₂₀ protein (n=30; P=0.033; Figure 11A and Figure 11D). In samples from 12 of 36 patients, a very weak or no caldesmon signal was detected (Figure 11C (i and ii]). These patients were diagnosed with renal inflammation (pyonephrosis, pyelonephritis, and ureteral inflammation). Furthermore, in the same patient group, we found lower values of the eGFR than in patients with no overt renal inflammation (Figure 11C [iii]).

Figure 11. Caldesmon protein expression in human intrarenal arteries in advanced age and by secondary inflammation.

A, Original western blot showing immunoreactivity of the membranes with transferred total homogenates of human intrarenal arteries and further incubated with antibodies against caldesmon, α‐actin, and MLC₂₀. B–D, Statistical evaluation of the correlation of the expression of caldesmon, α‐actin, MLC₂₀, and MYPT1 with age in patients (n=30). Expression was normalized to GAPDH, used as a housekeeping protein, and expressed as an absolute ratio of the immunoreactive signal of the gene of interest toward the signal of GAPDH. Linear regression was analyzed using Spearman’s correlation coefficient (r), as values <0.05 were considered significant. E, Caldesmon immunoreactivity expression in arterial homogenates from samples from patients with kidney carcinoma compared with samples from patients with inflammation (pyoonephrosis, pyelonephritis). Expression, normalized to GAPDH, was used as a housekeeping protein and depicted as an absolute ratio, caldesmon/GAPDH, in the figure E [i]). (E [ii]) Statistical evaluation of caldesmon expression, normalized to GAPDH, used as housekeeping protein and depicted as absolute ratio caldesmon/GAPDH in the figure (E [iii]). Statistical comparison of the values of estimated glomerular filtration rate between the patient with carcinoma and inflammation. P values represent the results from the correlative analysis or impaired t test. eGFR indicates estimated glomerular filtration rate; h‐Caldesmon, high‐molecular‐weight caldesmon; l‐Caldesmon, low‐molecular‐weight caldesmon; and MLC₂₀–20‐kDa myosin light chain.

Discussion

The present study investigated the contribution of caldesmon to the tone regulation of renal preglomerular vasculature. We have found that caldesmon is expressed along the preglomerular arteries of murine and human kidneys and decreases with age. In isolated mouse IAs, caldesmon counteracts the ROCK/PAK1‐dependent contractile pathway. Inflammation, which is known to impair renal 30, 31, 32 was featured by lower caldesmon expression. Moreover, we found that eGFR was diminished in patients with pyonephrosis/pyelonephritis.

Molecular Mechanisms of Vascular Tone Regulation by Caldesmon

The discovery of the F‐actin and calmodulin‐binding protein, caldesmon, as an endogenous inhibitor of smooth muscle actomyosin ATPase activity in the early 1980s ³³ prompted several research groups to investigate its regulatory effect on smooth muscle contractile activity. They showed that caldesmon inhibits vascular and visceral smooth muscle tone (reviewed in G CJaP ¹⁸ and Morgan and Gangopadhyay ³⁴ ). It has been demonstrated that the protein inhibits smooth muscle tone due to the actin‐binding C‐terminus of the protein. ¹⁵ , ³⁵ Caldesmon is found in the actomyosin domain of smooth muscle, ³⁶ where it additionally may serve a structural role by tethering actin and myosin filaments. ³⁷ Meanwhile, it has also been demonstrated that downregulation of caldesmon in vascular smooth muscle by caldesmon antisense oligodeoxynucleotides ³⁸ or small interfering RNA ¹⁷ leads to an increase in basal levels of unstimulated contractile tone, while stimulated tone was attenuated. Genetic ablation of high‐molecular‐weight caldesmon resulted in impaired relaxation in adult aortas, ³⁹ whereas ablation of both isoforms was perinatally lethal and attenuated stimulated tone in fetal bladders and aortas. ¹⁶ Taken together, these results support the notion that caldesmon is involved in the regulation of smooth muscle tone.

Caldesmon is phosphorylated by several protein kinases, including p21‐activated protein kinase 3. ²² Its phosphorylation was proposed to regulate its activity in addition to Ca²⁺‐calmodulin. ⁴⁰ Thus, it has been shown that caldesmon phosphorylation is concomitant with the contractile effect of agonists such as endothelin‐1 or activators of protein kinase C, ⁴¹ , ⁴² supporting the view that the endothelin‐1/protein kinase C/caldesmon axis seems to play a crucial role in fine‐tuning smooth muscle contraction. Furthermore, incubation of skinned muscle fibers with p21‐activated protein kinase 3 induced a slow contraction in the absence of MLC₂₀ phosphorylation. ²² Another interesting aspect was that its phosphorylation was also detected in nonstimulated adult but not fetal cerebral arteries, ⁴³ suggesting that it might regulate basal myogenic tone in an age‐dependent manner. In parallel, it was reported that caldesmon phosphorylation is not sensitive to histamine stimulation or the presence of NO donors, ⁴⁴ although NO donor application was able to reduce tone completely or the agonist application to constrict vascular smooth muscles maximally. These data support the view that not all contractile stimuli that can phosphorylate caldesmon affect its ability to disinhibit ATPase activity of the actomyosin complex and also that phosphorylation is regulated by a specific phosphatase distinct from myosin light‐chain phosphatase, which is the main target of cGMP/protein kinase G–induced relaxation or sympathetic stimulation. ⁴⁵ Studies on skinned fibers additionally demonstrated that phosphorylation per se is not involved in the sensitization of rabbit vasculature ⁴⁶ or tenia coli to Ca²⁺‐dependent actomyosin activation. ²³ All these findings showed that caldesmon expression rather than its phosphorylation is the primary mechanism for the caldesmon inhibitory effect. They additionally suggested that caldesmon phosphorylation is not the sole determinant of force and perhaps requires additional parallel stimulation.

Despite the aforementioned controversies regarding the role of caldesmon phosphorylation and myosin binding, several other studies demonstrated that the expression pattern of the protein is an unequivocal hallmark of various physiologic and pathologic conditions. ¹⁷ , ³⁸ Physiologically, myometrial high‐molecular‐weight caldesmon expression as well as the fraction of phosphorylated protein rise during the last trimester of pregnancy. ⁴⁷ Caldesmon protein is known to be downregulated in vasospastic conditions in cerebral vasculature ⁴⁸ or aged coronary vascular tissue, ⁴⁹ pointing out the importance of the inhibitory action of the protein for tone regulation of the circulatory system. Moreover, in carotid arteries, the loss of caldesmon parallels the maladaptive response of the vascular intima on occlusive stimulation balloon injury. ⁵⁰ Interestingly, in the latter model, caldesmon reduction was not associated with a change in smooth muscle α‐actin, suggesting that the effect could be specifically attributed to the rapid reduction of caldesmon rather than to the change in actin expression. ⁵⁰ As previously mentioned, reduction of the protein by small interfering RNA leads to the development of tone of vascular smooth muscle. ¹⁷ In this regard, herein we report that the high‐molecular‐weight caldesmon expression drops with advanced age (Figure 11A and Figure 11B). Moreover, in the majority of samples obtained from patients with accompanying kidney inflammation, caldesmon immunoreactive signals were not detected or were less prominent than those from the patients with carcinoma. We also have found that in the same patient group, reduced expression of caldesmon protein corresponds with the drop in eGFR (Figure 11C [iii]). It is reminiscent of the findings of others that inflammatory diseases such as interstitial pneumonia reduce caldesmon expression in lung vasculature but not the α‐smooth muscle actin. ⁵¹ Altogether, these findings indicate that the expression pattern of caldesmon relates to at least 2 independent factors, namely, advanced age and inflammation.

Activation of caldesmon CALD1 gene transcription in vascular smooth muscle cells depends mainly on serum response factor, which binds to the CC(A/T)6GG motif (CArG box) of the CALD1 promoter, and 2 other transcription factors: GATA6 and Nkx homeobox factor. ⁵² Serum response factor–mediated transcription of CALD1 in smooth muscle cells needs myocardin as a cofactor as well. ²⁴ Additionally, p53 upregulates caldesmon transcription, which can correlate with increased caldesmon levels observed in some cancers. ⁵³ , ⁵⁴ Insulin‐like growth factor 1 was identified as a main activator of high‐molecular‐weight caldesmon expression in primary cultures of vascular smooth muscle cells via the PI3K‐Akt/protein kinase B signal pathway. ⁵⁵ However, the circulating amount of insulin‐like growth factor 1 declines in aging, reaching low levels in individuals aged >60 years. ⁵⁶ , ⁵⁷ This decline of insulin‐like growth factor 1 production may contribute to the age‐dependent decrease of caldesmon in human vascular smooth muscle cells and thus be a reason for hypertension. Additionally, caldesmon deficiency may create specific pathogenic conditions in some vascular beds, such as kidney circulation, leading to renal malperfusion and kidney insufficiency due to a sustained increase in vascular tone of arcuate, interlobular, and afferent arterioles. On the other hand, regulatory molecules, like platelet‐derived growth factor and epidermal growth factor, leading to dedifferentiation of vascular smooth muscle cells, may also contribute to a decline in caldesmon expression, parallel to an additional reduction of h‐caldesmon level due to its conversion into low‐molecular‐weight caldesmon by coordinated activation of ERK1/2 and p38 mitogen‐activated protein kinase. ⁵⁷ , ⁵⁸

Together with its inhibitory effect on the contractile apparatus, several lines of evidence reveal that caldesmon is involved in the regulation of various cellular functions. In specimens obtained from patients with diabetic nephropathy, the caldesmon gene was found among 460 differentially expressed genes regulating actin cytoskeleton reorganization. ⁵⁹ It is also well known that low‐molecular‐weight caldesmon promotes tumor invasiveness, a process mediated by glucocorticoid receptor expression. ⁶⁰ Furthermore, loss of caldesmon induces tight junction disruption. ⁶¹ In model systems, it has been additionally found that the protein plays the role of an integral component of the podosomal membrane, which migrates from stress fibers to the actin core of the podosomes. ⁶² As the polymerization of cortical actin is an intracellular event mediated by ROCK, ⁶³ this supports the observation of the current study linking the regulatory pathway of ROCK with caldesmon inhibitory function. Herein, we report that caldesmon reduction lowers the reactivity of IAs to ROCK inhibition (Figure 8). We observed that the relaxation induced by pharmacological inhibition of another downstream effector of the ROCK cascade, PAK1, was also attenuated in caldesmon^+/− heterozygotes (Figure 9). In line with this, we recently observed that in murine basilar arteries, the phosphorylation responsible for PAK1 activation, T423, is sensitive to pharmacological inhibition of ROCK. ²⁵ A growing body of evidence has demonstrated that the dynamics of cortical actin polymerization are also involved in the regulation of vascular smooth muscle tone ⁶⁴ and specifically in the myogenic responses of rat brain vasculature. ⁶⁵ , ⁶⁶ In addition, it is well known that force could also be regulated in a ROCK‐dependent manner by a small fraction of nonmuscle myosin localized on subcortical cellular structures. ⁶⁷ As caldesmon is an integral protein that binds both actin and myosin, the present study supports previous observations made in the fetal urinary bladder that the protein plays a role as a stabilizer of the actomyosin complex. ¹⁶ Thus, in heterozygous mice or under pathological conditions (inflammation, aging), lack or reduction of caldesmon content in cortical structures could be critical for stabilization of the cortical actomyosin complex, manifested by a dysregulation of the tone of preglomerular vessels.

Study Limitations

Independent of observed good concordance between the data on caldesmon expression in murine and patient preglomerular vasculature, an important limitation of the present study is that it lacks functional experiments, revealing whether the reduction of caldesmon reflects on vascular reactivity in patients and murine kidney function in vivo. We also compared intrarenal arteries from different kidney compartments, that is, from murine IAs and from patient kidneys’ arcuate and interlobular arteries. Another limitation of the study is that no sex‐based analysis was performed due to the relatively low number of animals/patients included in each group.

Clinical Relevance

Altogether, our findings suggest that age and inflammation‐dependent downregulation of caldesmon protein could be used as a marker for renal vascular damage. Caldesmon reduction might contribute to impaired glomerular filtration, caused by imbalanced reactivity of preglomerular vasculature. In addition, while our study was still in progress, a novel clinical trial demonstrated that in patients with diabetes treatment with sodium–glucose cotransporter 2 inhibitor resulted in reduced caldesmon protein fragments in urine. ⁶⁸ The authors proposed that in diabetes, the kidney damage is accompanied by an increased degradation and excretion of caldesmon cleavage product with the urine. These results are reminiscent of our current findings showing a reduction of caldesmon when kidney function declines with advanced age and especially with kidney inflammation (Figure 11). Altogether, this strongly suggests that the diminished caldesmon abundance caused by accelerated protein degradation could be one of the reasons for diabetic kidney disease.

Perspectives

Future studies with larger patient cohorts would allow a parallel exploration of vascular reactivity versus caldesmon expression in each sample. Larger patient cohorts will also allow patients’ classification in subgroups, depending on their sex and preoperative diagnosis, and performance of a detailed correlative analysis on expression/contractility relationships. Furthermore, a detailed transcriptomic analysis will uncover the clinical conditions responsible for the reduction caldesmon expression with advanced age or inflammatory diseases such as pyonephrosis or pyelonephritis.

Sources of Funding

This study is financed by the European Union–NextGenerationEU, through the National Recovery and Resilience Plan of the Republic of Bulgaria, Project Number BG‐RRP‐2.004‐0009‐C02 (project MUVE‐TEAM, research group RenEVA: Lubomir T. Lubomirov, Nikolai Evtimov, Kamelia Bratoeva, Anton B. Tonchev, Viktor V. Velyanov, and Nikola R. Lazarov) and intramural funding from the State of Brandenburg, Germany, to Olaf Grisk Funding was also provided by the Deutsche Forschungsgemeinschaft (German Research Foundation), Project Numbers 397 484 323 (TRR259/B09), 468 236 352, and 384 170 921 (FOR2722/C01) to G.S. and as Deutsche Forschungsgemeinschaft instrument grant, Project Number 449115727 to Brandenburg Medical School Theodor Fontane.

Disclosures

None.

Supporting information

Data S1

Tables S1–S3

Figures S1–S3