The diabetic proximal tubule: part of the problem, and part of the solution?

Am J Physiol Renal Physiol 307: F147–F148, 2014; doi:10.1152/ajprenal.00272.2014.

Editorial Focus

The diabetic proximal tubule: part of the problem, and part of the solution? Alan M. Weinstein Department of Physiology and Biophysics, Department of Medicine, Weill Medical College of Cornell University, New York, New York

glucose handling dates to the era of whole organ experiments, with the description of a renal glucose threshold (Tm) and osmotic diuresis (22). Micropuncture identified the proximal tubule as the locus of glucose reabsorption (3), and brush border vesicle preparations (1, 8) and cellular electrophysiology (4, 18) established the Na⫹ dependence of that luminal glucose flux. Careful analysis of vesicle kinetics suggested two glucose transporters, one highcapacity, low-affinity site and a second with high affinity and smaller fluxes (17). The low-affinity carrier localized to the outer cortical region, and the high-affinity carrier localized to outer medullary vesicles. Corresponding to the affinity difference was the determination of a 1:1 (glucose:Na⫹) stoichiometry of the cortical cotransporter (6) and a 1:2 stoichiometry of the high-affinity carrier (16). Perfusion of isolated tubule segments confirmed the high-capacity carrier in the proximal convoluted tubule and the high-affinity transporter in the proximal straight tubule, whose maximum flux was ⬃10 –15% of that of the convoluted segment (2). An important advance in renal glucose transport came with the cloning of the gene for the intestinal cotransporter SGLT1 (SLC5A1) (5). In situ hybridization localized SGLT1 to the S3 segment of the proximal tubule, precisely the site suggested by the kinetic data (9). Oocyte expression of SGLT1 enabled extensive electrophysiological investigation and formulation of a mathematical model of the transporter. This work revealed that solute binding affinity is asymmetric, comparing inside and outside of the carrier, and translocation of the empty carrier is the important rate-limiting step and sensitive to the transmembrane potential difference (14). In pursuit of the high-capacity Na⫹-glucose cotransporter, homology screening revealed the gene for SGLT2 (SLC5A2), expressed in the kidney, for which the stoichiometry is 1:1 and which in situ hybridization localized to the S1 segment of the proximal tubule (7). Kinetic studies confirmed that SGLT2 is the low-affinity, high-capacity system identified in brush border vesicles (23). Subsequent study of Na⫹-glucose cotransport became more biophysically oriented, until the recent development of safe and specific inhibitors of SGLT2 thrust the topic back into view of the larger biomedical community. These substances are related to phlorizin, the classic nonspecific Na⫹-glucose cotransport inhibitor, and comprise a growing list, which includes dapagliflozin, canagliflozin, empaglibflozin, ipragliflozin, and tofogliflozin. When given to humans, especially at times of hyperglycemia, these drugs produce glycosuria; when given to diabetic patients, the glycosuria enhances glycemic control and in a manner that promotes caloric loss, rather than weight gain. With the advent of these medications, a new PHYSIOLOGICAL INTEREST IN RENAL

Address for reprint requests and other correspondence: A. M. Weinstein, Dept. of Physiology and Biophysics, Weill Medical College of Cornell Univ., 1300 York Ave., New York, NY 10021 (e-mail: [email protected]. edu). http://www.ajprenal.org

contingent of scientists and physicians has taken an interest in renal glucose metabolism. One point of concern in the adoption of these drugs was the specter of osmotic diuresis, in which glycosuria would produce natriuresis and hypovolemia, but this has not happened. Indeed, the observation is that the drugs inhibit ⬍50% of the reabsorption of filtered glucose, and there has been little hypoglycemia with their use. This finding has been posed as a paradox, in view of the fact that SGLT2 is normally responsible for ⬎90% of renal glucose transport (10). In that regard, it must be remembered that the amount of glycosuria depends upon local reabsorption rates of both Na⫹ and glucose, tubule fluid flow rate, and the fact that SGLT2 and SGLT1 are situated in series, so that the overall inhibitor impact is difficult to intuit without the aide of a model. This is the context for the work of Nagata et al. (13), whose paper reports innovations across several levels of function, and in so doing, is truly systems oriented. At the most basic level, this group has cloned SGLT1 and SGLT2 from the cynomolgus monkey and expressed the transporter in COS-7 cells. In these cells, they determine inhibitor affinities and demonstrate the 1,000-fold specificity of tofogliflozin for SGLT2 relative to SGLT1. They then perform classic whole animal studies in the monkeys, in which glucose excretion is plotted as a function of the filtered glucose load. One difficulty in these experiments is to implement a method of data analysis, which can capture the inhibitor’s impact. Because tofogliflozin is a competitive inhibitor, its effect diminishes at the highest luminal glucose concentrations, so that the Tm glucose is little changed; thus, representing the titration curve as two line segments won’t do. What these workers do is fit their glucose titration curve to an exponential, with three unknown parameters: the threshold for the filtered load at which excretion begins (xb), the classic Tm, and an exponential coefficient for the transition to complete glucose excretion. With this formulation, they capture the drug’s effect as the “splay” or difference of the integrated glucose excretion over the range of filtered loads. Within this analytic framework, both the threshold and splay are demonstrated to be functions of the inhibitor dose. What is (understandably) missing from this analysis is a model of the proximal tubule bridging the gap from the COS-7 cell data, i.e., using inhibitor KI and luminal solute concentrations, to predict the three model coefficients. Nevertheless, the data for anchoring such a model, at the transporter level and the whole kidney, have been supplied. What makes this work worth broadcasting in an editorial focus is the prospect that this line of investigation stands at the threshold of treatments that may protect the diabetic kidney. Vallon and coworkers (19, 20) have advanced the hypothesis that glomerular hyperfiltration in the diabetic kidney is the result of a primary increase in proximal Na⫹ reabsorption, producing a decrease in distal fluid delivery, and thus a secondary activation of tubuloglomerular feedback. The pathophysiology underlying the primary increase in Na⫹ transport is

1931-857X/14 Copyright © 2014 the American Physiological Society

F147

Editorial Focus F148 uncertain, but Vallon has pointed to oxidative cell stress with episodic high cytosolic glucose. Deficiency of the natriuretic action of dopamine has been implicated in the diabetic kidney, due to both decreased generation of dopamine (15), and decreased dopamine receptor (D1) activity (11, 12). Finally, even in the absence of cell damage, the model proximal tubule shows enhanced net Na⫹ reabsorption and decreased distal delivery, when increases in ambient glucose are still insufficient to provoke osmotic diuresis (21). The causal chain from glomerular hyperfiltration to the appearance of proteinuria, to glomerular fibrosis, to renal failure has become established teaching, and is supported by the observation that interventions which mitigate the hyperfiltration (angiotensin-converting enzyme inhibition or angiotensin receptor block) can, in some circumstances, delay the progression of renal damage. In this regard, the SGLT2 inhibitors appear to come with the potential for blunting episodic increases in cellular glucose uptake, and thus mitigating glomerular hyperfiltration at a point upstream in the pathophysiological chain of events. GRANTS This investigation was supported by Public Health Service Grant R01-DK29857 from the National Institute of Arthritis, Diabetes, and Digestive, and Kidney Diseases. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors. AUTHOR CONTRIBUTIONS Author contributions: A.M.W. drafted manuscript; A.M.W. edited and revised manuscript; A.M.W. approved final version of manuscript. REFERENCES 1. Aronson PS, Sacktor B. The Na⫹ gradient-dependent transport of D-glucose in renal brush border membranes. J Biol Chem 250: 6032–6039, 1975. 2. Barfuss DW, Schafer JA. Differences in active and passive glucose transport along the proximal nephron. Am J Physiol Renal Fluid Electrolyte Physiol 240: F322–F332, 1981. 3. Frohnert PP, Hohmann B, Zweibel R, Baumann K. Free flow micropuncture studies of glucose transport in the rat nephron. Pflügers Arch 315: 66 –85, 1970. 4. Frömter E, Gessner K. Active transport potentials, membrane diffusion potentials and streaming potentials across rat kidney proximal tubule. Pflügers Arch 351: 85–98, 1974. 5. Hediger MA, Coady MJ, Ikeda TS, Wright EM. Expression cloning and cDNA sequencing of the Na⫹/glucose co-transporter. Nature 330: 379 –381, 1987. 6. Hopfer J, Groseclose R. The mechanism of Na⫹-dependent D-glucose transport. J Biol Chem 255: 4453–4462, 1980.

7. Kanai Y, Lee W, You G, Brown D, Hediger MA. The human kidney low affinity Na⫹/glucose cotransporter SGLT2. Delineation of the major renal reabsorptive mechanism for D-glucose. J Clin Invest 93: 397–404, 1994. 8. Kinne R, Murer H, Kinne-Saffran E, Thees M, Sachs G. Sugar transport by renal plasma membrane vesicles. J Membr Biol 21: 375–395, 1975. 9. Lee WS, Kanai Y, Wells RG, Hediger MA. The high affinity Na⫹/ glucose cotransporter. Re-evaluation of function and distribution of expression. J Biol Chem 269: 12032–12039, 1994. 10. Liu J, Lee T, DeFronzo RA. Why do SGLT2 inhibitors inhibit only 30 –50% of renal glucose reabsorption in humans. Diabetes 61: 2199 – 2204, 2012. 11. Marwaha A, Banday AA, Lokhandwala MF. Reduced renal dopamine D1 receptor function in streptozotocin-induced diabetic rats. Am J Physiol Renal Physiol 286: F451–F457, 2004. 12. Moreira-Rodrigues M, Quelhas-Santos J, Serrão P, Fernandes-Cerqueira C, Sampaio-Maia B, Pestana M. Glycaemic control with insulin prevents the reduced renal dopamine D1 receptor expression and function in streptozotocin-induced diabetes. Nephrol Dial Transplant 25: 2945– 2953, 2010. 13. Nagata T, Suzuki M, Fukazawa M, Honda K, Yamane M, Yoshida A, Azabu H, Kitamura H, Toyota N, Suzuki Y, Kawabe Y. Competitive inhibition of SGLT2 by tofogliflozin or phlorizin induces urinary glucose excretion through extending splay in cynomolgus monkeys. Am J Physiol Renal Physiol 306: F1520 –F1533, 2014. 14. Parent L, Supplisson S, Loo DDF, Wright EM. Electrogenic properties of the cloned Na⫹/glucose cotransporter. II. A transport model under nonrapid equilibrium conditions. J Membr Biol 125: 63–79, 1992. 15. Stenvinkel R, Saggar-Malik AK, Wahrenberg H, Diczfalusy U, Bolinder J, Alvestrand A. Impaired intrarenal dopamine production following intravenous sodium chloride infusion in Type 1 (insulin-dependent) diabetes mellitus. Diabetologia 34: 114 –118, 1991. 16. Turner RJ, Moran A. Stoichiometric studies of the renal outer cortical brush border membrane D-glucose transporter. J Membr Biol 67: 73–80, 1982. 17. Turner RJ, Silverman M. Sugar uptake into brush border vesicles from dog kidney. II. Kinetics. Biochim Biophys Acta 511: 470 –486, 1978. 18. Ullrich KJ, Rumrich G, Kloss S. Specificity and sodium dependence of the active sugar transport in the proximal convolution of the rat kidney. Pflügers Arch 351: 35–48, 1974. 19. Vallon V. The proximal tubule in the pathophysiology of the diabetic kidney. Am J Physiol Regul Integr Comp Physiol 300: R1009 –R1022, 2011. 20. Vallon V, Huang D, Deng A, Richter K, Blantz RC, Thomson S. Salt-sensitivity of proximal reabsorption alters macula densa salt and explains the paradoxical effect of dietary salt on glomerular filtration rate in diabetes mellitus. J Am Soc Nephrol 13: 1865–1871, 2002. 21. Weinstein AM. Osmotic diuresis in a mathematical model of the rat proximal tubule. Am J Physiol Renal Fluid Electrolyte Physiol 250: F874 –F884, 1986. 22. Wesson, LG Jr, Anslow WP Jr. Excretion of sodium and water during osmotic diuresis in the dog. Am J Physiol 153: 465–474, 1948. 23. You G, Lee W, Barros EJG, Kanai Y, Huo T, Khawaja S, Wells RG, Nigam SK, Hediger MA. Molecular characterization of Na⫹-coupled glucose transporters in adult and embryonic rat kidney. J Biol Chem 270: 29365–29371, 1995.

AJP-Renal Physiol • doi:10.1152/ajprenal.00272.2014 • www.ajprenal.org

Am J Physiol Renal Physiol 307: F149 –F158, 2014. First published June 4, 2014; doi:10.1152/ajprenal.00439.2013.

Assessment of renal functional maturation and injury in preterm neonates during the first month of life Lina Gubhaju,1* Megan R. Sutherland,2* Rosemary S. C. Horne,3 Alison Medhurst,4 Alison L. Kent,5 Andrew Ramsden,4 Lynette Moore,6 Gurmeet Singh,7 Wendy E. Hoy,8 and M. Jane Black2 1

Preventative Health, Baker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia; 2Department of Anatomy and Developmental Biology, Monash University, Clayton, Victoria, Australia; 3Ritchie Centre for Baby Health Research, Monash Institute of Medical Research, Clayton, Victoria, Australia; 4Monash Newborn, Monash Medical Centre, Clayton, Victoria, Australia; 5Department of Neonatology, Canberra Hospital, and the Australian National University Medical School, Canberra, Australian Capital Territory, Australia; 6Department of Surgical Pathology, South Australia Pathology, Women’s and Children’s Hospital, North Adelaide and the University of Adelaide, Adelaide, South Australia, Australia; 7Menzies School of Health Research and the Royal Darwin Hospital, Casuarina, Northern Territory, Australia; and 8Centre for Chronic Disease, University of Queensland, Brisbane, Queensland, Australia Submitted 2 August 2013; accepted in final form 28 May 2014

Gubhaju L, Sutherland MR, Horne RS, Medhurst A, Kent AL, Ramsden A, Moore L, Singh G, Hoy WE, Black MJ. Assessment of renal functional maturation and injury in preterm neonates during the first month of life. Am J Physiol Renal Physiol 307: F149–F158, 2014. First published June 4, 2014; doi:10.1152/ajprenal.00439.2013.—Worldwide, approximately 10% of neonates are born preterm. The majority of preterm neonates are born when the kidneys are still developing; therefore, during the early postnatal period renal function is likely reflective of renal immaturity and/or injury. This study evaluated glomerular and tubular function and urinary neutrophil gelatinase-associated lipocalin (NGAL; a marker of renal injury) in preterm neonates during the first month of life. Preterm and term infants were recruited from Monash Newborn (neonatal intensive care unit at Monash Medical Centre) and Jesse McPherson Private Hospital, respectively. Infants were grouped according to gestational age at birth: ⱕ28 wk (n ⫽ 33), 29 –31 wk (n ⫽ 44), 32–36 wk (n ⫽ 32), and term (ⱖ37 wk (n ⫽ 22)). Measures of glomerular and tubular function were assessed on postnatal days 3–7, 14, 21, and 28. Glomerular and tubular function was significantly affected by gestational age at birth, as well as by postnatal age. By postnatal day 28, creatinine clearance remained significantly lower among preterm neonates compared with term infants; however, sodium excretion was not significantly different. Pathological proteinuria and high urinary NGAL levels were observed in a number of neonates, which may be indicative of renal injury; however, there was no correlation between the two markers. Findings suggest that neonatal renal function is predominantly influenced by renal maturity, and there was high capacity for postnatal tubular maturation among preterm neonates. There is insufficient evidence to suggest that urinary NGAL is a useful marker of renal injury in the preterm neonate. renal development; preterm birth; renal injury; proteinuria RENAL FUNCTION IN THE PRETERM neonate is affected by renal immaturity and potential injury during the early postnatal period. At the time when the majority of preterm infants are born, renal development is still ongoing (48) and renal function is accordingly immature (15). Preterm neonates have been shown to have a low glomerular filtration rate (GFR) compared with term neonates, and the tubules excrete high amounts of

* L. Gubhaju and M. R Sutherland are joint first authors. Address for reprint requests and other correspondence: M. J. Black, Dept. of Anatomy and Developmental Biology, School of Biomedical Sciences, Monash Univ., Clayton, Victoria, 3800 Australia (e-mail: [email protected]). http://www.ajprenal.org

sodium (2). Furthermore, compared with babies born at term, preterm neonates may demonstrate a slower progression in renal functional maturation after birth (5, 14). Creatinine clearance (CCr) has been shown to be positively correlated with both gestational age and postnatal age (2, 5, 7–9, 12, 14, 21, 39, 47, 51, 56), while the fractional excretion of sodium (FENa) has been shown to be inversely correlated with gestational age (12) and postnatal age (2, 12, 13, 39, 47). Only a few studies to date have investigated the occurrence of proteinuria following preterm birth. Very low concentrations of both high-molecular-weight (HMW) and low-molecular-weight (LMW) proteins are normally present in the urine, due to the function of the glomerular filtration barrier and the reuptake of filtered proteins in the proximal tubule (28, 32). A limited number of previous studies have demonstrated a high variation in urine albumin levels between individual preterm neonates (7, 10), with the highest levels exhibited by those with a low gestational age at birth and those that are clinically unstable (3, 7, 10, 11, 52). Urinary levels of ␤2-microglobulin (␤2-M) have also been shown to be significantly greater in the preterm infant compared with term-born infants throughout the first month of life (2, 52, 53), and they decrease with increasing gestational and postnatal age (50). Proteinuria is known to be an important indicator of renal injury; however, to date it remains unclear whether the increased urinary protein levels reported in preterm neonates are associated with renal immaturity and/or acute renal injury. The preterm kidney is highly susceptible to injury in the neonatal period; acute kidney injury (AKI) is reported to occur in 8 –24% of preterm neonates admitted to the neonatal intensive care unit (19, 44) and is primarily prerenal in origin (6, 44). In a large study of preterm infants born in the United States and Puerto Rico, Walker et al. (55) examined the medical records of 66,526 neonates (born at ⱕ30-wk gestation); 4% of the neonates were diagnosed with renal dysfunction and/or renal failure. The predominant risk factors for impaired renal function were low gestational age and low birth weight. Further risk factors included postnatal medication administration (vasopressors, indomethacin, and antibiotics), postnatal illness (intraventricular hemorrhage, a patent ductus arteriosus, necrotising enterocolitis, culture positive sepsis), and also the use of high-frequency ventilation, male gender,


F149

F150

RENAL FUNCTION AND INJURY IN PRETERM NEONATES

and non-white race. Importantly, mortality rates were significantly higher in neonates with the diagnosis of renal dysfunction and/or renal failure (55). In addition, renal injury in the preterm neonate may be an antecedent to chronic renal disease (1). Therefore, the early diagnosis of AKI is paramount so that conservative management of AKI can be initiated in time to potentially prevent these long-term consequences. In this regard, urinary neutrophil gelatinase-associated lipocalin (NGAL) has recently been investigated as a potential biomarker of AKI in preterm neonates, with studies showing that NGAL levels strongly correlated with both gestational and postnatal age (20, 23) and are highest in neonates that are critically ill (23, 33). In this Australian-based study, we have examined renal function in preterm infants admitted to the neonatal intensive care unit at the Monash Medical Centre (a large tertiary level hospital, located in Melbourne, Australia); the current rate of preterm birth in the Australian population is 8.2% (24). The aims of the study were to 1) assess postnatal renal function from day 3 to day 28 in extremely preterm (ⱕ28 wk of gestation), very preterm (29 –31 wk of gestation), and moderately preterm (32–36 wk of gestation) neonates compared with term controls (37– 42 wk of gestation) by examining glomerular (CCr, and urine albumin) and tubular (FENa and urine ␤2-M) function; and 2) to determine whether urinary NGAL is a useful marker of renal injury in preterm neonates. MATERIALS AND METHODS

Ethics Statement Ethics approval for this study was obtained from the Southern Health Human Research Ethics Committee and the Monash University Standing Committee on Ethics in Research Involving Humans. Written informed parental consent was obtained for all participants in the study. Study Population Preterm neonates (⬍37 wk of gestation) admitted to the neonatal intensive care unit at Monash Medical Centre and term infants (37– 42 wk of gestation) born at Jessie MacPherson Private Hospital (Clayton, Victoria, Australia) without any congenital abnormalities, were eligible for the study. Between April 2008 and October 2011, 143 neonates were recruited into the study. Four neonates were excluded following the early withdrawal of parental consent. Eight preterm neonates were further excluded following transfer to other hospitals before postnatal day 7. Two extremely preterm neonates died before the study was completed. The remaining 129 neonates were stratified into four groups according to gestational age: group A (ⱕ28 wk of gestation; n ⫽ 33); group B (29 –31 wk of gestation; n ⫽ 44); group C (32–36 wk of gestation; n ⫽ 30); and group D (ⱖ37 wk of gestation; n ⫽ 22). Urine Collection Procedure Sanitary pads (Kotex; Kimberly-Clark, NSW, Australia) were placed within diapers to collect urine samples, a method that has been previously validated (18, 45). A diaper liner (Johnson’s Baby Nappy Liners; Johnson&Johnson Pacific, NSW, Australia) was also placed inside the diaper to filter out any feces. To estimate urine flow rate, all diapers inclusive of the pad and liner were weighed before and after use, and the time each diaper was put on the baby was recorded (38). In the case of missed voids or a heavily soiled diaper, a value of average urine output for the relevant time period (calculated from all other diapers within that 24-h period) was substituted for the missing value. Diapers were changed at the discretion of nursing staff and/or parents (ranging from every 8 h in the extremely preterm neonates to

⬍2 h in the term neonates), and collected in a sealed plastic container. At least twice per day, diapers were collected from the nursery and the urine was extracted by compressing the sanitary pad using a hydraulic press (18, 45). All urine collected from the diapers over a 24-h period was pooled before analysis. It has been shown that urine collected from disposable cotton pads and/or cotton wool does not affect the urinary constituent of sodium, potassium, or creatinine (38) and has been used previously in the analysis of urinary NGAL (23). Previous research has shown that protein can get bound within cotton material (43). Therefore, spot urine samples collected using urine collection bags were utilized to determine urine total protein, albumin, and ␤2-microglobulin levels. In preterm neonates, 24-h urine collection from diapers began at 72 h after birth (day 3) and continued until postnatal day 7. In addition, 24-h urine was collected on days 14, 21, and 28 of life. Spot urine samples (1–2 ml) were obtained on days 7, 14, 21, and 28 of life. In term neonates, 24-h urine collection commenced 48 h after birth (day 2) and continued until the infant was discharged from the hospital (⬃day 4 of life). Additionally, urine was collected for a 24-h period on day 28 of life. For those infants who had been discharged, diapers were delivered and collected from the infant’s home (group A, n ⫽ 0; group B, n ⫽ 0; group C, n ⫽ 7; and group D, n ⫽ 11). Spot urine samples were obtained from the term neonates on day 3 of life. Pooled 24-h urine samples were frozen at ⫺20°C until analysis. Spot urine samples were sent for analysis of urinary protein levels immediately after collection. For a number of the infants, the urine collections and analyses on days 14 –28 were not performed, primarily due to the transfer of infants to private hospitals, or discharge from the hospital; group A [n ⫽ 1/33 (3.0%)]; group B [n ⫽ 8/44 (18.2%)]; group C [n ⫽ 11/30 (33.3%)]; and group D [n ⫽ 11/22 (50.0%)]. Assessment of Renal Function Urinary and plasma sodium and creatinine. All urine analyses were performed by the Southern Health Pathology Department (Southern Cross Pathology; Clayton, Victoria, Australia). Pooled 24-h urine samples were analyzed for sodium and creatinine levels. Plasma creatinine and plasma sodium levels were recorded from blood tests that were undertaken as part of the routine care of preterm neonates, and data were extracted from the medical records. In term neonates, heel-prick blood samples were obtained for analysis at the time of the routine newborn screening test (at ⬃48 h of life). In cases where a blood test was not available in term infants on postnatal day 28 (group D, n ⫽ 11), average levels of plasma sodium (140 mmol/l) and plasma creatinine (40 ␮mol/l) that are within the expected range for term infants were used in the calculation of CCr and FENa. Calculation of CCr and FENa. The estimation of GFR using CCr is generally considered to be a more accurate measure of glomerular function than serum creatinine alone. Although it is to be noted that the potential for active tubular secretion of creatinine, as well as differences in muscle mass between neonates, may affect the accuracy of the results (46). CCr was calculated using the following formula: CCr 共ml/min/BSA兲 ⫽ 共UCr/PCr兲 * urine flow rate/BSA, where UCr ⫽ urinary creatinine (␮mol/l); urine flow rate ⫽ ml/min (calculated for each 24-h period); PCr ⫽ plasma creatinine (␮mol/l); and BSA ⫽ body surface area (m2). BSA was calculated using the following formula derived by Haycock et al. (17):

共BSA兲共m2兲 ⫽ 0.024265 * body weight 共kg兲0.5378

* body length 共cm兲0.3964.

FENa was calculated using the following formula:



FENa共%兲 ⫽ 共UNa/PNa兲 * 共PCr/Ucr兲 * 100, where UNa ⫽ urinary sodium (mmol/l); PNa ⫽ plasma sodium (mmol/l); PCr ⫽ plasma creatinine (␮mol/l); and UCr ⫽ urinary creatinine (␮mol/l). Urinary total protein, albumin, and ␤2-M. Urine total protein (UTP), urinary albumin, and urinary ␤2-M were measured in spot urine samples using nephelometric technology on a Beckman immunochemistry system, with reagents and calibrators supplied by Beckman Diagnostics (urine total protein and urine albumin; Beckman Diagnostics; Sydney, Australia) and DakoCytomation (␤2-M; DakoCytomation; Glostrup, Denmark). In instances where UTP was greater than 500 mg/l, this was defined as pathological proteinuria and urinary albumin levels were not determined. All urine protein levels were expressed as a ratio to urine creatinine concentrations. It is to be noted that missing data exist for the urinary protein results due to difficulties in obtaining clean spot urine samples from a number of the infants (urine collection bags could not be placed on some extremely preterm neonates due to their delicate skin and/or parents not providing consent). Overall, 46.2% of the total number of requested spot urine samples were obtained. At least one spot urine sample was obtained for the majority of neonates: group A (91%), group B (93%), group C (87%), and group D (27%). Urinary NGAL. In a subset of the study participants (61.2%), pooled 24-h urine samples from one or more postnatal time points were analyzed for urinary NGAL levels. Urinary NGAL analysis was performed using a sandwich ELISA in microwells coated with a monoclonal antibody against human NGAL (NGAL ELISA Kit, BioPorto Diagnostics; Gentofte, Denmark); the upper limit of the test was 500 ng/ml. In accordance with previous studies, urinary NGAL levels were expressed as nanograms per milliliter and were not corrected for urine creatinine concentrations (20, 23, 33). Impaired renal function. Impaired renal function (low urine output, low creatinine clearance, high serum creatinine, high fractional excretion of sodium, high urine total protein and high urinary NGAL) was defined as values that differed more than 2 SD from the mean at any time point from postnatal day 3 through to postnatal day 28. The mean and SD was calculated from either absolute values (SCr) or natural log (urine output, serum Cr, FENa, UTP, NGAL) or squareroot (CCr) transformed values (transformations were performed to ensure the normality of data), inclusive of all preterm neonates (groups A, B, and C). To assess the relationship between the six measures of renal dysfunction and whether any preterm neonate exhibited multiple measures of renal impairment, the total number of measures of renal dysfunction for each neonate was also examined. Statistical Analysis Statistical analyses were performed using GraphPad Prism v5.04 for Windows and Intercooled Stata v8.0 for Windows. Data are presented as the means ⫾ SE. Statistical significance was accepted at the level of P ⬍ 0.05. Birth characteristics (gestational age, birth weight, length, head circumference) were compared among groups using a one-way ANOVA, followed by a Bonferroni post hoc test. To determine differences in categorical variables among groups (such as sex, disease outcomes, and medication administration), a Fisher’s exact test was performed. Urine output, CCr, and FENa in preterm neonates (groups A, B, and C) were analyzed using a two-way ANOVA with repeated measures to assess renal functional maturation with increasing postnatal age. At postnatal days 3 and 28, urine output, CCr and the FENa (in all 4 groups) were analyzed using a two-way ANOVA, followed by a Bonferroni post hoc test to determine differences between individual groups at each time point; this analysis enabled comparison between preterm and term infants at the start and end points of the study. Urine protein (UTP, albumin, and ␤2-M)-to-creatinine ratios and urine

F151

NGAL levels were also analyzed using a two-way ANOVA, followed by a Bonferroni post hoc test. The factors assessed in all of these analyses were gestational age (pGA), postnatal age (pPA), and their interaction (pGA ⫻ PA). Linear regression analyses [followed by an analysis of covariance (ANCOVA)] were used to compare the rate of change in CCr and FENa from postnatal day 3 to day 28, and also to determine whether urinary NGAL levels correlated with any other indices of renal function. Additionally, urinary NGAL levels in neonates exhibiting pathological proteinuria (UTP ⱖ500 mg/l) were compared with ageand sex-matched neonates (controls) with lower urine total protein levels (UTP ⬍480 mg/l, and less than half of the UTP level of the matched neonate with pathological proteinuria) at the corresponding postnatal time point. This analysis was undertaken using a two-way ANOVA with the factors proteinuria (pP), gestational age (pGA), and their interaction (pP ⫻ GA). Additionally, in these two groups, urinary NGAL levels were assessed at time points before (7 days if pathological proteinuria observed at day 14, 21, or 28; 3 days if observed at day 7) and at the time of proteinuria onset. This analysis was undertaken using a two-way ANOVA with the factors proteinuria (pP), time point of assessment (pT), and their interaction (pP ⫻ T). RESULTS

Pregnancy and Neonatal Birth Characteristics Reasons for preterm delivery included onset of spontaneous preterm labor (34.0%), premature prelabor rupture of membranes (30.1%), placenta previa/abruption (15.5%), and clinical indication due to maternal and/or fetal health risks such as preeclampsia and suspected fetal compromise (20.4%). There were a significantly higher proportion of births attributed to spontaneous preterm labor in group A (extremely preterm) neonates compared with both group B and group C neonates. The majority of preterm neonates (⬎65%) were born via caesarean delivery, compared with 27% among term infants (Table 1). At least 70% of the mothers of preterm neonates received antenatal steroids before delivery; 94% of those born extremely preterm (group A) received antenatal steroids. Body weight, length, and head circumference at birth all increased significantly along with increasing gestational age (Table 1). There was no significant difference in the gender balance between gestational age groups. Groups A and C had the greatest number of small-for-gestational age (SGA) neonates; there were a similar number of multiple births in each of the preterm groups whereas all term-born neonates in group D were singletons. Of the preterm groups, there was no significant difference in birth weight between the singletons and multiples (group A: singleton 818.6 ⫾ 36.8 g, multiple 790.9 ⫾ 35.8 g; group B: singleton 1,429 ⫾ 47.9 g, multiple 1,468 ⫾ 65.0 g; group C: singleton 1,767 ⫾ 111.4 g, multiple 1,778 ⫾ 93.4 g). The majority of neonates that had a low Apgar score (ⱕ7) at 5 min were in group A (Table 2). Postnatal Neonatal Complications and Medications Administered Since the majority of preterm neonates required mechanical ventilation after birth, respiratory distress syndrome was very common in these groups (Table 2). The occurrence of culturepositive sepsis and a patent ductus arteriosus was significantly greater in group A compared with those infants in older gestational age groups. Only neonates within group A and group B were diagnosed with intraventricular hemorrhage.


F152


Table 1. Pregnancy and birth characteristics of neonates by gestational age group Gestational age, wk Birth weight, g Body length, cm Head circumference, cm Male, % SGA, % Twin/triplet, % Antenatal steroids Caesarean

Group A, ⱕ28 wk (n ⫽ 33)

Group B, 29–31 wk (n ⫽ 44)

Group C, 32–36 wk (n ⫽ 30)

Group D, ⱖ37 wk (n ⫽ 22)

26.6 ⫾ 0.2* (24–28) 811.1 ⫾ 28.3* (529–1,229) 33.6 ⫾ 0.4* (29–39) 23.9 ⫾ 0.2* (21–27) 36.4 30.3†§ 27.3† 93.8† 60.6†

30.4 ⫾ 0.1* (29–31) 1,438.0 ⫾ 39.6* (892–2,157) 40.9 ⫾ 0.4* (36–46) 28.3 ⫾ 0.2* (25–32) 52.3 9.1‡¶ 22.7† 94.7† 67.4†

33.7 ⫾ 0.2* (32–36) 1,771.0 ⫾ 75.4* (1,018–2,542) 43.4 ⫾ 0.7* (36–50) 29.9 ⫾ 0.4* (26–33) 46.7 46.7†§ 40.0† 80.8† 64.3†

39.6 ⫾ 0.2* (37–42) 3,356.0 ⫾ 76.2* (2,820–4,160) 49.9 ⫾ 0.4* (46–55) 34.0 ⫾ 0.3* (32–36) 50.0 4.5‡¶ 0.0* 0.0* 28.6*

Values are means ⫾ SE (range) and as noted. SGA, small for gestational age. Significant differences (P ⬍ 0.05) between groups are indicated by symbols: *vs. all other groups; †vs. group D; ‡vs. group C; §vs. group B; ¶vs. group A.

Among preterm neonates, the most commonly administered medications were a routine regimen of antibiotics. The majority of neonates that received additional antibiotics were in the extremely preterm group (group A). In general, the administration of drugs was significantly greater among neonates in group A. Term infants (group D) did not receive any medications over the course of the study. Renal Function Urine output. Neither gestational age at birth nor postnatal age affected urine output in preterm neonates from postnatal day 3 to day 7 (Fig. 1A). On postnatal days 14 –28, however, urine output was significantly greater in neonates with increased gestational age at birth (Fig. 1B). Urine output was significantly lower in term neonates compared with preterm neonates at postnatal day 3; however, there was no difference between groups at postnatal day 28 (Fig. 1C). Oliguria (urine output ⬍1 ml·kg⫺1·h⫺1) was observed in 10 term infants (postnatal day 3) and 1 preterm neonate (postnatal day 5). CCr. During the first week and month of life, CCr was positively associated with both gestational age at birth and postnatal age (Fig. 1, D and E). Of the preterm neonates,

groups B and C had the highest CCr throughout the study period. On postnatal day 28, Ccr levels were not significantly different among the preterm groups; however, term neonates had significantly higher Ccr compared with preterm neonates (Fig. 1F). Linear regression analyses showed that there was no significant difference in the rate of change in CCr from postnatal day 3 to postnatal day 28 among all groups of neonates (data not shown). FENa. FENa was inversely associated with gestational age at birth during the first week and month of life (Fig. 1, G and H). The highest levels of sodium excretion were observed in group A throughout the study period. On postnatal day 3 (Fig. 1I), FENa was significantly lower in term neonates (group D) than in all other groups. By postnatal day 28, however, there was no significant difference in FENa among groups. Linear regression analysis showed that the rate of change in FENa from postnatal day 3 to day 28 was significantly greater in group B neonates compared with group D (P ⬍ 0.0001) and was also significantly greater in group A neonates compared with all other groups (P ⱕ 0.02; data not shown). UTP, albumin, and ␤2-M. UTP, albumin, and ␤2-M levels in preterm neonates (corrected for urine creatinine) were all

Table 2. Percentage of neonates with postnatal complications and percentage exposed to medications during the early postnatal period in each gestational age group

Postnatal complications Apgar ⱕ7 at 5 min Mechanical ventilation Respiratory distress syndrome Culture positive sepsis Intraventricular hemorrhage Patent ductus arteriosus Antibiotics ␤-Lactam (benzylpenicillin, imipenem, ampicillin) Aminoglycoside (gentamicin) Glycopeptide (vancomycin) Macrolide (erythromycin) Cephalosporin (cefotaxime, cefozolin) Nitroimidazole (metronidazole) Other medications Antifungal (nilstat, nystatin, fluconazole) Methylxanthine (aminophylline, theophylline) Inotrope (dopamine, dobutamine) Steroid (hydrocortisone) Diuretic (furosemide) NSAID (indomethacin)

Group A, ⱕ28 wk (n ⫽ 33) (%)

Group B, 29–31 wk (n ⫽ 44) (%)

Group C, 32–36 wk (n ⫽ 30) (%)

Group D, ⱖ37 wk (n ⫽ 22) (%)

45.5* 100†‡ 100†‡ 50.0* 31.3†‡ 68.8*

25.6* 84.1†‡ 87.2† 12.2¶ 14.6 15.0¶

3.3§¶ 56.7* 50.0† 3.3¶ 0.0¶ 3.3¶

0.0§¶ 0.0* 0.0* 0.0¶ 0.0¶ 0.0¶

80.0* 80.0* 10.0¶ 3.3 3.3 3.3

0.0* 0.0* 0.0¶ 0.0 0.0 0.0

14.3§¶ 17.9 3.3¶ 0.0¶ 3.3¶ 0.0¶

0.0§¶ 0.0§¶ 0.0¶ 0.0¶ 0.0¶ 0.0¶

100†‡ 100†‡ 57.6* 12.5 12.5 3.1 56.3†‡ 37.5†‡ 28.1* 3.1* 34.4* 34.4*

100†‡ 100†‡ 16.7¶ 2.4 2.4 0.0 46.2†‡ 29.3† 0.0¶ 0.0¶ 7.7¶ 2.3¶

Significant differences (P ⬍ 0.05) between groups are indicated by symbols: *vs. all other groups; †vs. group D; ‡vs. group C; §vs. group B; ¶vs. group A. AJP-Renal Physiol • doi:10.1152/ajprenal.00439.2013 • www.ajprenal.org


F153

Fig. 1. Urine output (A–C), creatinine clearance (D–F), and the fractional excretion of sodium (G–I) in neonates born at ⱕ28 wk of gestation (group A), 29 –31 wk of gestation (group B), 32–36 wk of gestation (group C), and ⱖ37 wk of gestation (group D) on postnatal days 3–7 (top row; group A, n ⫽ 33; group B, n ⫽ 44; group C, n ⫽ 29); days 14 –28 (middle row; group A, n ⫽ 23; group B, n ⫽ 19; group C, n ⫽ 7); and (bottom row) day 3 (group A, n ⫽ 34; group B, n ⫽ 44; group C, n ⫽ 30; group D, n ⫽ 20) vs. day 28 (group A, n ⫽ 32; group B, n ⫽ 24; group C, n ⫽ 9; group D, n ⫽ 11). Data were analyzed using a 2-way ANOVA (with repeated measures for line graphs) with the factors gestational age (pGA), postnatal age (pPA), and their interaction (pGAxPA). From Bonferroni post hoc analysis: *P ⬍ 0.01 compared with groups A, B, and C (C and F). P ⬍ 0.05, a vs. b and c, b vs. c (F and I).

inversely associated with gestational age at birth (Fig. 2). There was no change in urine protein levels with increasing postnatal age. There was wide intragroup variability in urine protein levels; within group A for example, UTP:Cr levels ranged from 92.2 to 759.3 mg/mmol at postnatal day 28 (Fig. 2A). In term infants, UTP:Cr (mean: 219.3 ⫾ 139.9 mg/mmol) and albumin:Cr (mean: 19.9 ⫾ 9.1 mg/mmol) levels measured on postnatal day 3 were similar to the levels found among group B neonates at postnatal day 7; ␤2-M:Cr levels (mean: 0.8 ⫾ 0.2 mg/mmol), however, were negligible. Pathological proteinuria (UTP ⱖ500 mg/l) was observed in 12 (9.3%) neonates, at one or more postnatal time points, with the majority of these in group A (7/12), followed by group B (2/12), group C (1/12), and group D (2/12). Urinary NGAL. There was a significant effect of gestational age on urinary NGAL, with the lowest NGAL levels observed at postnatal day 28 in group D term neonates (Fig. 3A). Nine preterm neonates (7.0%) had levels of urinary

NGAL ⬎390 ng/ml (⬎2 SD from the mean) at one or more postnatal time points, with the majority (7/9) occurring on postnatal day 28. There was a significant positive linear correlation between urinary NGAL levels and UTP:Cr (Fig. 3B); however, urinary NGAL levels were not correlated with serum creatinine, CCr, FENa, urine albumin, or ␤2-M levels (data not shown). Urinary NGAL levels in the subset of preterm neonates who exhibited pathological proteinuria (UTP ⱖ500 mg/l) were not different from the NGAL levels of age- and sex-matched controls with low protein excretion at corresponding postnatal time points (Fig. 3C). In these two groups, with the exception of a few neonates, urinary NGAL levels were not predictive of proteinuria (Fig. 3D); urinary NGAL levels were not significantly increased at the time points before or at the onset of pathological proteinuria. Impaired renal function. Overall, ⬃14% of preterm neonates from each gestational age group were classified as having low

Fig. 2. Urine total protein (UTP; A), albumin (B), and ␤2-microglobulin (␤2-M; C) levels corrected for urine creatinine, on postnatal days 7, 14, 21, and 28 in preterm neonates (groups A, n ⫽ 33; B, n ⫽ 40; and C, n ⫽ 25). Data were analyzed using a 2-way ANOVA with the factors pGA, pPA, and pGAxPA. AJP-Renal Physiol • doi:10.1152/ajprenal.00439.2013 • www.ajprenal.org

F154


Fig. 3. Urine neutrophil gelatinase-associated lipocalin (NGAL) levels (A) in neonates born at ⱕ28 wk of gestation (group A; n ⫽ 25), 29 –31 wk of gestation (group B; n ⫽ 20), 32–36 wk of gestation (group C; n ⫽ 11), and ⱖ37 wk of gestation (group D; n ⫽ 22) at postnatal days 3 and 28. Data were analyzed using a 2-way ANOVA with the factors pGA, pPA, and pGAxPA. Linear regression analysis is shown of urinary NGAL vs. UTP/creatinine ratio in preterm neonates (n ⫽ 68, B). Urinary NGAL (ng/ml) levels were also assessed in preterm neonates with pathological proteinuria (UTP ⱖ500 mg/l; grey; n ⫽ 12), with values compared with age- and sex-matched controls with normal UTP at the corresponding time point (black; n ⫽ 12), grouped by gestational age (C). Groups were compared using a 2-way ANOVA, with the factors pGA, proteinuria (pP), and their interaction (pGAxP). To determine whether NGAL is predictive of proteinuria (D), urinary NGAL levels were assessed in the preterm neonates with pathological proteinuria (grey; n ⫽ 12) compared with age- and sex-matched controls (black; n ⫽ 12) at time points before and at the time of pathological proteinuria onset. The factors assessed were the time point of assessment (pT), pP, and their interaction (PTxP).

urine output (Fig. 4A). Twenty-five percent of neonates in group A, 9.5% in group B, and 7.1% in group C had high serum creatinine levels (⬎100.6 ␮mol/l); only one neonate (group B), however, was categorized as having low CCr (Fig. 4, B and C). High FENa was predominantly observed in group A neonates, with three preterm neonates from groups A and B exhibiting hyponatremia (serum sodium levels ⬍130 mmol/l) during the study period (Fig. 4D). High UTP levels were only observed in group B and group C neonates, whereas high urinary NGAL levels were observed in a small percentage of neonates in each gestational age group (Fig. 4, E and F). The relationship between the six different measures of renal dysfunction (described in Fig. 4) and whether any preterm neonates had multiple measures of renal impairment were also examined. Twelve (11.2%) preterm neonates had more than two measures of renal impairment. As shown in Table 3, the most common combination of renal dysfunction measures exhibited during the first month of life was high serum creatinine and high FENa (observed in 7 of 12 neonates). DISCUSSION

The findings of this study demonstrate that renal function in preterm neonates, during the first month of life, is significantly affected by gestational age at birth and postnatal age. Together, these results suggest that neonatal renal function is predominantly influenced by renal structural maturity. By postnatal day 28, CCr was significantly lower among preterm neonates compared with term infants; however, differences in FENa were not observed, which is suggestive of a high capacity for postnatal

tubular maturation. Both urinary protein and NGAL levels were inversely associated with gestational age at birth, which suggests that they are markers of renal immaturity. Of concern, pathological proteinuria was observed in 12 preterm neonates; among those neonates, urinary NGAL levels were not elevated, which has two potential implications: 1) that the cause of the pathological proteinuria is not due to acute kidney injury (as NGAL has previously been shown to be a marker of AKI); or 2) if the pathological proteinuria observed is due to renal injury, urinary NGAL is not a useful marker or predictor of renal injury in this population. The findings from this study provide normative values for a number of renal functional parameters in preterm infants. The presence of renal dysfunction in a high proportion of babies highlights the importance of conducting large-scale studies among populations of preterm neonates and the close monitoring of renal function in preterm infants in the neonatal intensive care unit (NICU). Maturation of Renal Function in the Early Postnatal Period In this study, urine output was not significantly different during the first week of life among preterm neonates. This finding was expected, as within the NICU fluid intake is strictly maintained according to neonatal body weight, and from approximately postnatal day 4 urine output is known to be predominantly influenced by fluid intake (26, 27). From postnatal day 14 to postnatal day 28, a significantly higher urine output was observed among neonates born at older gestational ages, which likely relates to the change in feeding regime. All term infants were exclusively breast-fed in the first week after


F155


Fig. 4. Percentage of neonates with impaired renal function (grey) and those with adequate renal function (black) grouped by gestational age (group A; n ⫽ 33; group B; n ⫽ 44; group C; n ⫽ 30, and group D; n ⫽ 22). Impaired renal function (values that differed by ⬎2 SD from the mean at any postnatal age, days 3–28) was indicated by low urine output (A), low creatinine clearance (CCr; B), high serum creatinine (C), high fractional excretion of sodium (FENa; D), high urine total protein (E), or high urinary NGAL (F).

birth; hence their observed low urine output is likely due to low maternal milk production within the first few days after birth. Since this was a noninvasive study, urine output was estimated using diaper weights; it is important to note that this method has its limitations since estimates based on averages had to be used in occasional instances of missed voids and soiled diapers. The 24-h urine collections conducted at home were also highly reliant on the record-keeping of parents, and this noninvasive technique was therefore prone to error. Furthermore, the number of infants assessed after postnatal day 7 was reduced due to loss to follow-up; therefore, the robustness of data was diminished for all analyses conducted on postnatal days 14 –28. Consistent with previous studies, as gestational age at birth and postnatal age increased, CCr increased (2, 5, 7–9, 12, 14, 21, 39, 47, 51, 56) and FENa decreased (2, 12, 13,

39, 47). Although neonates born very or extremely preterm commence with a low GFR and high FENa, by postnatal day 28 both CCr and FENa were not significantly different between the preterm gestational age groups, which is suggestive of a high capacity for postnatal glomerular and tubular maturation among those born very or extremely preterm. Interestingly, the rate of change in CCr over the first month of life was very similar between all groups of neonates, and in contrast to previous studies (5, 14), a slower postnatal increase in CCr in preterm neonates compared with term infants was not observed. In extremely preterm neonates, there was a significantly greater rate of change in FENa from postnatal day 3 to day 28 compared with all other groups, suggestive of more accelerated tubular maturation. In support of this idea, at postnatal day 28 CCr remained significantly lower among

Table 3. Measures of renal dysfunction in 12 preterm neonates that exhibited multiple (ⱖ2) measures of renal dysfunction, by gestational age grouping and sex Gestational Age Group

Sex

Low Urine Output

A (ⱕ28 wk) A (ⱕ28 wk) A (ⱕ28 wk) A (ⱕ28 wk) A (ⱕ28 wk) B (29–31 wk) B (29–31 wk) B (29–31 wk) B (29–31 wk) C (32–36 wk) C (32–36 wk) C (32–36 wk)

F F F M M M F M M M F M

X

Low CCr

High Serum Cr

High FENa

X X X X

X X X X X X X

X X X X

High UTP

High NGAL

X

X X

X X X X

X X

CCr, creatinine clearance; FENa, fractional excretion of sodium; NGAL, neutrophil gelatinase-associated lipocalin. AJP-Renal Physiol • doi:10.1152/ajprenal.00439.2013 • www.ajprenal.org

X

F156


preterm neonates compared with term infants whereas there was no significant difference in FENa. Overall, these results suggest that renal functional capacity increases with renal structural maturity, potentially through increased filtration surface area (increased number of nephrons in those neonates with ongoing postnatal nephrogenesis (48) and/or increased capillary growth) (31, 48), the maturation of tubular cells (40), as well as the substantial changes in renal blood flow and renal vascular resistance that occur after birth (54). In accordance with previous studies (2, 3, 7, 10, 11, 52, 53), we observed that urine albumin, ␤2-M, and UTP levels were inversely associated with gestational age at birth. There was no change, however, in urine protein levels with increasing postnatal age. Proteinuria in the preterm neonate may reflect either structural immaturity or injury to the glomerular filtration barrier (allowing for the passage of albumin into the filtrate), and/or impaired uptake of filtered protein in the proximal tubule. Unlike the significant reduction in sodium excretion with increasing maturity, the capacity for protein reabsorption remained low postnatally, which possibly reflects a much slower maturation of renal protein handling compared with sodium handling. Alternatively, these findings may be indicative of glomerular or tubular injury in the preterm neonate; proximal tubule cell injury (such as occurs following oxidative stress in the preterm neonate) (36) and/or an overload of filtered protein are possible causes of impaired tubular protein uptake. Evidence of Renal Dysfunction in Preterm Neonates In this study, renal impairment was defined as measures of renal function that were ⬎2 SD from the mean. There is a current lack of definition in the literature as to what constitutes AKI in the preterm neonate; if the general definition of AKI commonly used in adults [RIFLE criteria (4, 30)] was applied in the current study, only one preterm neonate (and 10 term neonates presenting with oliguria) clearly met the criteria. Therefore, we adopted a broader definition of dysfunction in the current study to give an indication as to the percentage of infants with reduced renal functional capacity (glomerular and tubular), rather than focusing on the strict AKI criteria. In the present study, 25% of neonates in group A, and ⬍10% in groups B and C, exhibited high serum creatinine levels (⬎2 SD from the mean); there was just one preterm neonate (group B), however, that had low CCr levels (from days 3– 6 of life). A relatively high percentage of preterm neonates, predominantly those born extremely preterm, were observed to have a high percentage of sodium excretion; however, likely due to the administration of sodium supplementation in this population, the majority of preterm neonates maintained adequate levels of serum sodium. It may be speculated that the high sodium excretion in these neonates represents either tubular immaturity or injury. Of the neonates that exhibited more than one measure of renal dysfunction, the majority had both high FENa and high serum creatinine. If tubular function is impaired, this may also have an impact on serum creatinine levels as creatinine (besides predominantly being filtered by the glomeruli) is known to be actively excreted by proximal tubular cells (25, 35), and in the case of preterm neonates, may also be reabsorbed by the immature tubules as has been observed in a neonatal animal model (29). Serum creatinine levels are also

influenced by extrarenal factors such as muscle mass, and the intake of nitrogen, protein, and creatinine (25, 35) which is increased with milk formula and other parenteral nutrition preparations (37). Each of these factors may have influenced the relatively high proportion of neonates who also exhibited high serum creatinine levels; in contrast, only one preterm neonate was found to have low CCr. In general, CCr is considered to be a more reliable indicator of GFR than serum creatinine levels alone (46), especially given the large number of factors that may influence the generation of creatinine as described above. Encouragingly, this finding may indicate that the capacity for glomerular filtration in the preterm kidney in the early neonatal period is quite adequate; however, tubular function is likely impaired. Pathological proteinuria was observed among 12 neonates in this study (including 2 term-born infants) at 1 or more postnatal time points. In this regard, the only preterm neonate with low CCr also exhibited pathological proteinuria on postnatal days 14, 21, and 28. Interestingly, the onset of severe proteinuria was often observed at a later stage (9/12 neonates exhibited pathological proteinuria after day 21), which suggests that factors in the postnatal clinical course (rather than renal immaturity) may be the cause of the severe proteinuria. In support of this idea, it was only neonates born at older gestational ages that exhibited high UTP levels (⬎2 SD from the mean). Exposure to nephrotoxic medications (including NSAIDs and antibiotics) are known to impair nephrogenesis (42, 49) and cause renal injury, such as podocyte foot process effacement (22). It is important to note that with regard to all the maternal and fetal factors we examined (including medications), there was no direct correlation with renal impairment, thus indicating that the renal dysfunction is likely to be multifactorial in origin. Future studies in a larger cohort of neonates are required to identify individual factors (such as exposure to medications and growth restriction) which may be contributing to renal dysfunction following preterm birth. Urinary NGAL as a Marker of Acute Kidney Injury NGAL is excreted by renal proximal tubule cells as a response to AKI (41); however, NGAL is produced during nephrogenesis (16), and levels may also be raised in late-onset sepsis (34). Although positive findings have been reported among older infants (57), the usefulness of urinary NGAL as a marker of AKI in preterm neonates remains unclear (20, 23, 33, 34). Certainly, consistent with previous studies (20, 23) we found a significant inverse correlation between urinary NGAL and gestational age at birth; these findings may relate to the immaturity of the kidney, and the clinical instability of the younger neonates. Importantly, the results of the current study demonstrated that urinary NGAL levels were not directly correlated with any indicators of renal dysfunction, apart from UTP. Although the correlation between NGAL and UTP was statistically significant, it is to be noted that the coefficient of determination was very low. Furthermore, only two preterm neonates were observed to have concurrently high UTP and NGAL levels; in general, neonates with pathological proteinuria did not exhibit high urinary NGAL levels. There was also no association between NGAL levels at time points before or at the time of



proteinuria onset, suggesting that NGAL may not be predictive of renal injury and/or proteinura. Conclusions

5. 6.

The findings of this study demonstrate that renal maturity is an important determinant of glomerular and tubular function among preterm neonates. Of particular concern, a number of preterm neonates exhibited severe proteinuria; however, there was no correlation with urinary NGAL levels. Given the immaturity of the preterm kidney, it is important to determine what levels of protein in the urine are normal vs. those that are pathological and thus indicative of renal injury in preterm neonates. Furthermore, it would be of value to identify specific factors in the postnatal clinical care of the preterm neonate which may be leading to the high urinary protein excretion. The results from this study have been important in working toward the development of a normal range of urinary protein levels in preterm neonates, but our findings are limited by the relatively small sample size and the large inter- and intragroup variability observed. The consequences of proteinuria in the neonatal period are unknown; however, the potential for progressive renal injury and long-term renal dysfunction suggests the need for regular assessments of renal function in subjects that are born preterm.

7. 8.

9. 10.

11. 12.

13.

14.

ACKNOWLEDGMENTS The authors acknowledge the valuable assistance of the following people: all medical, nursing, and administrative staff at Monash Newborn NICU and the Jessie MacPherson maternity ward at the Monash Medical Centre; nursing staff at Dandenong Hospital and Casey Hospital special care nurseries; Kom Yin, who assisted with studies in the term neonates; Michael Daskalakis and staff at the Southern Health Pathology Department; Susan Mott for assistance with statistical analysis; and all of the families that participated in the study, especially those that undertook diaper collections in their home.

15. 16.

17.

18.

GRANTS This project was supported by a National Health and Medical Research Council of Australia project grant. L. Gubhaju and M. R. Sutherland were recipients of Australian Postgraduate Awards.

19. 20.

DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors.

21.

AUTHOR CONTRIBUTIONS Author contributions: L.G., R.S.C.H., A.M., A.L.K., A.R., L.M., G.S., W.E.H., and M.J.B. provided conception and design of research; L.G. and M.R.S. performed experiments; L.G. and M.R.S. analyzed data; L.G., M.R.S., R.S.C.H., A.M., A.L.K., A.R., L.M., G.S., W.E.H., and M.J.B. interpreted results of experiments; L.G. and M.R.S. prepared figures; L.G. and M.R.S. drafted manuscript; L.G., M.R.S., and M.J.B. edited and revised manuscript; L.G., M.R.S., R.S.C.H., A.M., A.L.K., A.R., L.M., G.S., W.E.H., and M.J.B. approved final version of manuscript.

22.

23.

24.

REFERENCES 1. Abitbol CL, Bauer CR, Montane B, Chandar J, Duara S, Zilleruelo G. Long-term follow-up of extremely low birth weight infants with neonatal renal failure. Pediatr Nephrol 18: 887–893, 2003. 2. Aperia A, Broberger O, Elinder G, Herin P, Zetterstrom R. Postnatal development of renal function in pre-term and full-term infants. Acta Paediatr Scand 70: 183–187, 1981. 3. Awad H, el-Safty I, el-Barbary M, Imam S. Evaluation of renal glomerular and tubular functional and structural integrity in neonates. Am J Med Sci 324: 261–266, 2002. 4. Bellomo R, Ronco C, Kellum J, Mehta R, Palevsky P. Acute renal failure— definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Con-

25. 26.

27.

28. 29.

F157

ference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 8: R204 –R212, 2004. Bueva A, Guignard J. Renal function in the preterm neonate. Pedatr Res 36: 572–577, 1994. Cataldi L, Leone R, Moretti U, De Mitri B, Fanos V, Ruggeri L, Sabatino G, Torcasio F, Zanardo V, Attardo G, Riccobene F, Martano C, Benini D, Cuzzolin L. Potential risk factors for the development of acute renal failure in preterm newborn infants: a case-control study. Arch Dis Child Fetal Neonatal Ed 90: F514 –F519, 2005. Clark PM, Bryant TN, Hall MA, Lowes JA, Rowe DJ. Neonatal renal function assessment. Arch Dis Child 64: 1264 –1269, 1989. Cuzzolin L, Fanos V, Pinna B, di Marzio M, Perin M, Tramontozzi P, Tonetto P, Cataldi L. Postnatal renal function in preterm newborns: a role of diseases, drugs and therapeutic interventions. Pediatr Nephrol 21: 931–938, 2006. Fawer CL, Torrado A, Guignard JP. Maturation of renal function in full-term and premature neonates. Helv Paediatr Acta 34: 11–21, 1979. Fell JM, Thakkar H, Newman DJ, Price CP. Measurement of albumin and low molecular weight proteins in the urine of newborn infants using a cotton wool ball collection method. Acta Paediatr 86: 518 –522, 1997. Galaske RG. Renal functional maturation: renal handling of proteins by mature and immature newborns. Eur J Pediatr 145: 368 –371, 1986. Gallini F, Maggio L, Romagnoli C, Marrocco G, Tortorolo G. Progression of renal function in preterm neonates with gestational age ⬍ or ⫽ 32 weeks. Pediatr Nephrol 15: 119 –124, 2000. Giapros V, Papadimitriou P, Challa A, Andronikou S. The effect of intrauterine growth retardation on renal function in the first two months of life. Nephrol Dial Transplant 22: 96 –103, 2007. Gordjani N, Burghard R, Leititis J, Brandis M. Serum creatinine and creatinine clearance in healthy neonates and prematures during the first 10 days of life. Eur J Pediatr 148: 143–145, 1998. Gubhaju L, Sutherland MR, Black MJ. Preterm birth and the kidney: implications for long-term renal health. Reprod Sci 18: 322–333, 2011. Gwira JA, Wei F, Ishibe S, Ueland JM, Barasch J, Cantley LG. Expression of neutrophil gelatinase-associated lipocalin regulates epithelial morphogenesis in vitro. J Biol Chem 280: 7875–7882, 2005. Haycock G, Schwartz G, Wisotsky D. Geometric method for measuring body surface area: a height-weight formula validated in infants, children and adults. J Pediatr 93: 62–66, 1978. Heckmann M, Hartmann MF, Kampschulte B, Gack H, Bodeker RH, Gortner L, Wudy SA. Assessing cortisol production in preterm infants: do not dispose of the nappies. Pediatr Res 57: 412–418, 2005. Hentschel R, Lodige B, Bulla M. Renal insufficiency in the neonatal period. Clin Nephrol 46: 54 –58, 1996. Huynh TK, Bateman DA, Parravicini E, Lorenz JM, Nemerofsky SL, Sise ME, Bowman TM, Polesana E, Barasch JM. Reference values of urinary neutrophil gelatinase-associated lipocalin in very low birth weight infants. Pediatr Res 66: 528 –532, 2009. Iacobelli S, Bonsante F, Ferdinus C, Labenne M, Gouyon JB. Factors affecting postnatal changes in serum creatinine in preterm infants with gestational age ⬍32 weeks. J Perinatol 29: 232–236, 2009. Kent AL, Maxwell LE, Koina ME, Falk MC, Willenborg D, Dahlstrom JE. Renal glomeruli and tubular injury following indomethacin, ibuprofen, and gentamicin exposure in a neonatal rat model. Pediatr Res 62: 307–312, 2007. Lavery AP, Meinzen-Derr JK, Anderson E, Ma Q, Bennett MR, Devarajan P, Schibler KR. Urinary NGAL in premature infants. Pediatr Res 64: 423–428, 2008. Laws PJ, Li Z, Sullivan EA. Australia’s Mothers and Babies 2008. Perinatal Statistics Series Number 24. Canberra, Australia: Ausralian Institute of Health and Welfare, 2010. Levey A, Perrone RD, Madias N. Serum creatinine and renal function. Annu Rev Med 39: 465–490, 1988. Lorenz JM, Kleinman LI, Ahmed G, Markarian K. Phases of fluid and electrolyte homeostasis in the extremely low birth weight infant. Pediatrics 96: 484 –489, 1995. Lorenz JM, Kleinman LI, Kotagal UR, Reller MD. Water balance in very low-birth-weight infants: relationship to water and sodium intake and effect on outcome. J Pediatr 101: 423–432, 1982. Mathieson P. The cellular basis of albuminuria. Clin Sci 107: 533–538, 2004. Matos P, Duarte-Silva M, Drukker A, Guignard J. Creatinine reabsorption by the newborn rabbit kidney. Pediatr Res 44: 639 –641, 1998.


F158


30. Mehta R, Kellum J, Shah S, Molitoris B, Ronco C, Warnock D, Levin A. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 11: R31, 2007. 31. Moore L, Williams R, Staples A. Glomerular dimensions in children under 16 years of age. J Pathol 171: 145–150, 1993. 32. Mundel P, Reiser J. Proteinuria: an enzymatic disease of the podocyte? Kidney Int 77: 571–580, 2010. 33. Parravicini E. The clinical utility of urinary neutrophil gelatinase-associated lipocalin in the neonatal ICU. Curr Opin Pediatr 22: 146 –150, 2010. 34. Parravicini E, Nemerofsky SL, Michelson KA, Huynh TK, Sise ME, Bateman DA, Lorenz JM, Barasch JM. Urinary neutrophil gelatinaseassociated lipocalin is a promising biomarker for late onset culturepositive sepsis in very low birth weight infants. Pediatr Res 67: 636 –640, 2010. 35. Perrone R, Madias N, Levey A. Serum creatinine as an index of renal function: new insights into old concepts. Clin Chem 38: 1933–1953, 1992. 36. Perrone S, Mussap M, Longini M, Fanos V, Bellieni CV, Proietti F, Cataldi L, Buonocore G. Oxidative kidney damage in preterm newborns during perinatal period. Clin Biochem 40: 656 –660, 2007. 37. Premji S, Fenton T, Sauve R. Higher versus lower protein intake in formula-fed low birth weight infants. Cochrane Database Syst Rev CD003959: 2006. 38. Roberts S, Lucas A. Measurement of urinary constituents and output using disposable napkins. Arch Dis Child 60: 1021–1024, 1985. 39. Ross B, Cowett R, Oh W. Renal functions of low birth weight infants during the first two months of life. Pediatr Res 11: 1162–1164, 1977. 40. Satlin L, Woda C, Schwartz G. Development of function in the metanephric kidney. In: The Kidney: From Normal Development to Congenital Disease, edited by Vize P, Woolf A, and Bard J. Sydney, Australia: Academic, 2003. 41. Schmidt-Ott KM, Mori K, Li JY, Kalandadze A, Cohen DJ, Devarajan P, Barasch J. Dual action of neutrophil gelatinase-associated lipocalin. J Am Soc Nephrol 18: 407–413, 2007. 42. Schreuder MF, Bueters RR, Huigen MC, Russel FG, Masereeuw R, van den Heuvel LP. Effect of drugs on renal development. Clin J Am Soc Nephrol 6: 212–217, 2011. 43. Smith GC, Taylor CM. Recovery of protein from urine specimens collected in cotton wool. Arch Dis Child 67: 1486 –1487, 1992. 44. Stapleton F, Jones D, Green R. Acute renal failure in neonates: Incidence, etiology and outcome. Pediatr Nephrol 1: 314 –320, 1987.

45. Stark MJ, Clifton VL, Wright IM. Carbon monoxide is a significant mediator of cardiovascular status following preterm birth. Pediatrics 124: 277–284, 2009. 46. Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney function—measured and estimated glomerular filtration rate. N Engl J Med 354: 2473–2483, 2006. 47. Sulyok E, Varg F, Gyory E, Jobst K, Czaba I. Postnatal development of renal sodium handling in premature infants. J Pediatr 95: 787–792, 1979. 48. Sutherland MR, Gubhaju L, Moore L, Kent AL, Dahlstrom JE, Horne RS, Hoy WE, Bertram JF, Black MJ. Accelerated maturation and abnormal morphology in the preterm neonatal kidney. J Am Soc Nephrol 22: 1365–1374, 2011. 49. Sutherland MR, Yoder BA, McCurnin D, Seidner S, Gubhaju L, Clyman RI, Black MJ. Effects of ibuprofen treatment on the developing preterm baboon kidney. Am J Physiol Renal Physiol 302: F1286 –F1292, 2012. 50. Takieddine F, Tabbara M, Hall P, Sokol RJ, King KC. Fetal renal maturation. Studies on urinary beta 2 microglobulin the neonate. Acta Obstet Gynecol Scand 62: 311–314, 1983. 51. Thayyil S, Sheik S, Kempley ST, Sinha A. A gestation- and postnatal age-based reference chart for assessing renal function in extremely premature infants. J Perinatol 28: 226 –229, 2008. 52. Tsukahara H, Fujii Y, Tsuchida S, Hiraoka M, Morikawa K, Haruki S, Sudo M. Renal handling of albumin and beta-2-microglobulin in neonates. Nephron 68: 212–216, 1994. 53. Tsukahara H, Yoshimoto M, Saito M, Sakaguchi T, Mitsuyoshi I, Hayashi S, Nakamura K, Kikuchi K, Sudo M. Assessment of tubular function in neonates using urinary beta 2-microglobulin. Pediatr Nephrol 4: 512–514, 1990. 54. Veille JC, McNeil S, Hanson R, Smith N. Renal hemodynamics: longitudinal study from the late fetal life to one year of age. J Matern Fetal Investig 8: 6 –10, 1998. 55. Walker MW, Clark RH, Spitzer AR. Elevation in plasma creatinine and renal failure in premature neonates without major anomalies: terminology, occurrence and factors associated with increased risk. J Perinatol 31: 199 –205, 2011. 56. Wilkins BH. Renal function in sick very low birthweight infants. 1. Glomerular filtration rate. Arch Dis Child 67: 1140 –1145, 1992. 57. Zappitelli M, Washburn KK, Arikan AA, Loftis L, Ma Q, Devarajan P, Parikh CR, Goldstein SL. Urine neutrophil gelatinase-associated lipocalin is an early marker of acute kidney injury in critically ill children: a prospective cohort study. Crit Care 11: R84, 2007.


Am J Physiol Renal Physiol 307: F159 –F171, 2014. First published May 28, 2014; doi:10.1152/ajprenal.00546.2013.

RhoA/Rho kinase mediates TGF-␤1-induced kidney myofibroblast activation through Poldip2/Nox4-derived reactive oxygen species Nagaraj Manickam,2* Mandakini Patel,2* Kathy K. Griendling,3 Yves Gorin,2 and Jeffrey L. Barnes1,2 1

The Medical Research Service, Audie Murphy Memorial Veterans Administration Hospital, South Texas Veterans Health Care System, San Antonio, Texas; 2The Department of Medicine, Division of Nephrology, The University of Texas Health Science Center, San Antonio, Texas; and 3The Department of Medicine, Division of Cardiology, Emory University School of Medicine, Atlanta, Georgia Submitted 18 October 2013; accepted in final form 18 May 2014

Manickam N, Patel M, Griendling KK, Gorin Y, Barnes JL. RhoA/Rho kinase mediates TGF-␤1-induced kidney myofibroblast activation through Poldip2/Nox4-derived reactive oxygen species. Am J Physiol Renal Physiol 307: F159 –F171, 2014. First published May 28, 2014; doi:10.1152/ajprenal.00546.2013.—The small G proteins Rac1 and RhoA regulate actin cytoskeleton, cell shape, adhesion, migration, and proliferation. Recent studies in our laboratory have shown that NADPH oxidase Nox4-derived ROS are involved in transforming growth factor (TGF)-␤1-induced rat kidney myofibroblast differentiation assessed by the acquisition of an ␣-smooth muscle actin (␣-SMA) phenotype and expression of an alternatively spliced fibronectin variant (Fn-EIIIA). Rac1 and RhoA are essential in signaling by some Nox homologs, but their role as effectors of Nox4 in kidney myofibroblast differentiation is not known. In the present study, we explored a link among Rac1 and RhoA and Nox4-dependent ROS generation in TGF-␤1-induced kidney myofibroblast activation. TGF-␤1 stimulated an increase in Nox4 protein expression, NADPH oxidase activity, and abundant ␣-SMA and Fn-EIIIA expression. RhoA but not Rac1 was involved in TGF-␤1 induction of Nox4 signaling of kidney myofibroblast activation. TGF-␤1 stimulated active RhoA-GTP and increased Rho kinase (ROCK). Inhibition of RhoA with small interfering RNA and ROCK using Y-27632 significantly reduced TGF-␤1-induced stimulation of Nox4 protein, NADPH oxidase activity, and ␣-SMA and Fn-EIIIA expression. Treatment with diphenyleneiodonium, an inhibitor of NADPH oxidase, did not decrease RhoA activation but inhibited TGF-␤1-induced ␣-SMA and Fn-EIIIA expression, indicating that RhoA is upstream of ROS generation. RhoA/ROCK also regulated polymerase (DNA-directed) ␦-interacting protein 2 (Poldip2), a newly discovered Nox4 enhancer protein. Collectively, these data indicate that RhoA/ROCK is upstream of Poldip2-dependent Nox4 regulation and ROS production and induces redox signaling of kidney myofibroblast activation and may broader implications in the pathophysiology of renal fibrosis. fibrosis; NADPH oxidase; oxidative stress; GTPase; myofibroblast differentiation; Rho kinase; transforming growth factor-␤1; polymerase (DNA-directed) ␦-interacting protein 2

(TGF)-␤1 is the predominant growth factor responsible for mesenchymal cell synthesis of the matrix during fibrosis (3, 14, 46). A hallmark of mesenchymal cell activation is the acquisition of a myofibroblast phenotype, whereby fibroblasts acquire the expression of smooth muscle proteins, most notably ␣-smooth muscle actin (␣-SMA), and synthesize the extracellular matrix. TGF-␤1 regulates differentia-

TRANSFORMING GROWTH FACTOR

* N. Manickam and M. Patel contributed equally to this work. Address for reprint requests and other correspondence: J. L. Barnes, Div. of Nephrology, Dept. of Medicine, The Univ. of Texas Health Science Center, 7703 Floyd Curl Dr., San Antonio, TX 78229-3900 (e-mail: barnesj@uthscsa. edu). http://www.ajprenal.org

tion of fibroblasts to a myofibroblast phenotype (3, 11, 54) and subsequent matrix synthesis, particularly of the alternatively spliced isoform of fibronectin (Fn-EIIIA) (3, 6, 54). The signaling pathways for TGF-␤1-induced myofibroblast differentiation are complex and not completely understood. The fibroblast-to-myofibroblast transition is differentially regulated by the cononical Smad pathway (28). In mesangial cells and fibroblasts, TGF-␤/Smad signaling (Smad2/3) is tightly controlled by MAPK (Ras/MEK/ERK) signaling cascades (3, 24, 66). Also, ERK and Akt/PKB act as alternative pathways in TGF-␤1 signaling of matrix proteins (15, 24, 25, 30), making these three signaling proteins important transduction molecules in myofibroblast differentiation. ROS are also known to be second messengers in cell activation (19). Recent studies have indicated that NADPH oxidase-generated ROS, particularly from the Nox4 homolog, appears to be responsible for TGF-␤ signaling of myofibroblast activation and differentiation (3, 10). We have recently observed TGF-␤1-induced myofibroblast differentiation and expression of Fn-EIIIA signals through a redox pathway involving Nox4 and ERK in kidney fibroblasts (5). A variety of studies also have indicated that a small G protein signaling pathway involving Rho family GTPases (RhoA and Rac1) act as molecular switches for a variety of cellular functions that could be involved in myofibroblast differentiation (29). Of these, RhoA and its downstream effector Rho kinase (ROCK) are known to regulate myofibroblast differentiation (26, 50). The RhoA/ROCK pathway constitutes an important mediator for a variety of cell functions, including actin cytoskeleton reorganization, regulation of cell shape, adhesion, migration, and proliferation (26, 51). As with TGF-␤1 and Nox4, a close association exists between RhoA/ ROCK and myofibroblast activation by regulating the actin cytoskeleton, stress fiber formation, and ␣-SMA expression in progenitor cells, pericytes, and fibroblasts (20, 26, 47, 51). RhoA/ROCK may also be involved in the regulation of NADPH oxidase signaling (39, 53, 64); however, if or where RhoA/ROCK functions in TGF-␤1 signaling of renal myofibroblast differentiation is not known. Also, the small G protein Rac1 is necessary for Nox1 and Nox2 activation and is a component of the active enzymatic complex; however, control of Nox4 function through Rac1 is not clear and remains controversial (3, 8, 18). Nox4 does not require the classical cytoplasmic subunits or Rac1, and activation of Nox4 is undisturbed in cells where Rac1 is knocked down, supporting the notion that Nox4 is Rac independent (41) and suggesting a role for other G protein transduction pathways, perhaps through RhoA. In addition, polymerase (DNA-directed) ␦-interacting protein 2 (Poldip2), a Nox4 F159

F160

KIDNEY MYOFIBROBLAST ACTIVATION VIA THE RhoA/ROCK/Poldip2/Nox4 PATHWAY

enhancing protein, induces cytoskeletal changes, including strengthening of focal adhesions and stress fiber formation in vascular smooth muscle cells, suggesting interplay among these signaling proteins (39, 57). A role for RhoA and Poldip2 interactions with Nox4 has not been explored in TGF-␤1-induced myofibroblast activation. The present study investigated a role for Rac1 and RhoA/ROCK in Poldip2/Nox4 signaling in TGF-␤1induced kidney myofibroblast differentiation and cell migration. MATERIALS AND METHODS

Kidney fibroblast cell culture. Normal rat kidney fibroblast cells (NRK-49F) were obtained from the American Type Culture Collection (ATCC; Rockville, MD). Cells were grown in RPMI medium containing 10% FCS, antibiotics, and antimycotics in a humidified incubator at 37°C under 5% CO2. Cells were grown to 80% confluence and then rendered quiescent by replacing the complete medium with serum-free medium for a period of 24 h, as previously described (5). Quiescent cells were treated with TGF-␤1 (1 ng/ml) and then analyzed for Rac1, RhoA/ROCK, and NADPH oxidase signaling pathways. In experiments using chemical inhibitors of the various signaling pathways, compounds were added 0.5–2 h before the addition of TGF-␤1 and incubated until the time of harvest. The following inhibitors were used: NSC-23766 (50 ␮M) for Rac1 (Calbiochem, EMD Biosciences, San Diego, CA), small interfering (si)RNA targeting RhoA (siRhoA; Dharmacon Products, Thermo Scientific, Lafayette, CO), ROCK inhibitor Y-27632 (10 ␮M, Calbiochem), NADPH oxidase inhibitor diphenyleneiodonium (100 nM, Sigma-Aldrich, St. Louis, MO), and siRNA targeting Poldip2 (siPoldp2; Life Technologies-Invitrogen, Grand Island, NY). Previous studies have indicated that Nox4 signaling is an early event, occurring within 5 min of stimulation by TGF-␤1, possibly due to translational mechanisms (5); thus, readout times for signaling proteins Rac1, RhoA, Nox4, and Poldip2 ranged from 5 min to 6 h. Fibroblast-to-myofibroblast differentiation was assessed by the expression of ␣-SMA and Fn-EIIIA detected by immunoblot analysis and immunohistochemistry at 24 or 48 h after stimulation with TGF-␤1. Adenoviruses. NH2-terminal Myc-tagged Poldip2 (AdPoldip2) (39) or active wild-type Nox4 (40) and adenovirus with empty vector or noninfected fibroblasts served as controls. NRK-49F cells were infected for 1 h with recombinant adenovirus in serum-free culture medium and incubated for another 2 days in the same medium without virus before being harvested for protein extraction. Western blot analysis. Immunoblot analysis was performed as previously described (5, 13). Cells were lysed in RIPA buffer and centrifuged at 4°C at 14,000 rpm for 5 min, and the supernatant was retained and quantified for protein using the Bio-Rad DC protein assay method (Bio-Rad Laboratories, Herculus, CA). After samples had been boiled in sample buffer for 10 min, equal amounts of protein lysate were loaded onto SDS-PAGE gels and electrophoretically separated. Separated proteins were transferred to polyvinylidene difluoride membranes followed by blockade with 5% nonfat dry milk. Protein bands were detected by enhanced chemiluminescence using standard ECL detection methods as recommended by the manufacturer (Amersham Pharmacia Biotech). Signals were visualized by film radiography or using a Bio-Rad Chemidoc XRS photo documentation system (Bio-Rad Laboratories) and represented by column graphs as average intensity percentages of TGF-␤1 intensity. The antibodies used were targeted against ␣-SMA (clone 1A4, Sigma Chemical), Fn-EIIIA (clone IST-9, Abcam, Cambridge, MA), Nox4 [H-300, Santa Cruz Biotechnology, Santa Cruz, CA, or produced in our laboratory (17)], Poldip2 (Santa Cruz Biotechnology), Rac1 (Cytoskeleton, Denver, CO), RhoA (Millipore, Temecula, CA), ROCK, and phosphorylated myosin phosphatase targeting protein-1 (MYPT-1; Santa Cruz Biotechnology). Western blot data were normalized to GAPDH (anti-GAPDH, Novus Biologicals, Littelton, CO) or total nonphosphorylated protein.

Rac1 and RhoA activation assays. Activation of Rac1 and RhoA was determined using GTP pulldown assay kits according to the manufacturers’ instructions (Rac1: Cytoskeleton and RhoA: Millipore). After cells were lysed, active GTP-Rac1 was pulled down from 150 ␮g of cleared lysate using 10 ␮g of PAK-PBD beads. Similarly, active GTP-RhoA was bound using 15 ␮g of glutathione-agarose bound glutathione-S-transferase-tagged rhotekin-RBD. For both assays, samples were incubated at 4°C on a rotator for 1 h, washed three times in PBS, resuspended in Laemmli buffer, and then resolved with 12.5% SDS-PAGE. Lysates and precipitates were analyzed by immunoblot analysis using mouse monoclonal antibody specifically against Rac1 or RhoA. ROS detection by 2=,7=-dichlorodihydrofluorescein assay. Kidney fibroblasts were cultured in a multiwell LAB-TEK no. 1 coverslide chambers (Nalge Nunc, Naperville, IL) and then made quiescent at 30 – 40% confluence as described above. To explore a role for RhoA/ ROCK in endogenous ROS generation, cells were treated with Y-27632 (10 ␮M) 1 h before TGF-␤1 stimulation, and 2=,7=-dichlorodihydrofluorescein (DCF) fluorescence was examined as an indicator of ROS generation, as previously described (5, 17). The peroxidesensitive fluorescent probe DCF diacetate (10 ␮M, Molecular Probes, Eugene, OR) was added to cells under various conditions, and photographs were taken 30 min after the addition of TGF-␤1 using an Olympus Fluoview confocal laser scanning inverted microscope equipped with a 40A planfluor objective. DCF fluorescence was detected with an excitation wavelength of 488 nm, and DCF emission was detected using a 510- to 550-nm band-pass filter. NADPH oxidase assay. NADPH oxidase was measured by the lucigenin-enhanced chemiluminescence method as previously described (5, 17). Cells were treated with NSC-23766, Y-27632, or vehicle and then stimulated with TGF-␤1 for 30 min. Cells were washed in PBS and homogenized in lysis buffer containing 20 mM KH2PO4 (pH 7.0), 1 mM EGTA, 1 mM PMSF, 10 ␮g/ml aprotinin, and 0.5 ␮g/ml leupeptin. Homogenate (100 ␮l) was added to 900 ␮l of 50 mM potassium phosphate buffer (pH 7.0) containing 1 mM EGTA, 150 mM sucrose, 5 ␮M lucigenin (Sigma Chemical) as an electron acceptor, and 100 ␮M NADPH (Roche Diagnostics, Indianapolis, IN) as an electron donor. Photon emission expressed as relative light units was measured every 30 s for 5 min in a Modulus luminometer (Turner BioSystems, Promega Biosystems, Sunnyvale, CA). There was no measurable activity in the absence of NADPH. Protein content was measured using Bio-Rad protein assay reagent. NADPH oxidase activity was expressed as the rate of relative chemiluminescence units per milligram of protein per minute. H2O2 is the major product of Nox4, and thus measuring superoxide underestimates its activity. Immunofluorescence microscopy. Immunofluorescence microscopy was used to examine the relative roles of Rac1 and RhoA/ROCK in myofibroblast ␣-SMA and Fn-EIIIA expression. Quiescent kidney fibroblast cells grown in multiwell plastic chamber slides were treated with NSC-23766 or Y-27632 followed by TGF-␤1 and then incubated for 24 or 48 h. At the termination of the study time, cells were washed twice with ice-cold PBS and fixed in methanol at ⫺20°C for 5 min. After a brief rinse, cells were blocked with 0.1% BSA in PBS and then stained with Cy3-conjugated mouse anti-␣-SMA. Fn-EIIIA was detected using indirect immunofluorescence as previously described (5). Cy3-conjugated donkey anti-mouse served as secondary antibody (Chemicon and Millipore). Stained cells were washed with PBS, mounted with coverslips, viewed by epifluorescence, and photographed using an Olympus AX70 Research microscope equipped with a DP-70 digital camera (Melville, NY). siRNA transfection. To specifically examine a role for RhoA in kidney fibroblast ROS generation, cells were transfected with siGenome smart pool rat RhoA or a nontargeting RNA pool as a control using DharmaFECT 2 transfection agent (Thermo/Dharmacon Research, Lafayette, CO). Likewise, a role for Poldip2 in kidney fibroblast activation was examined using Stealth siRNA targeting rat Poldip2. Nontargeting siRNA served as a control. Cells were grown to 60% confluence in six-well culture plates, and the medium was then



replaced with antibiotic/antimycotic-free RPMI containing siRNA in DharmaFECT 2 reagent at a concentration of 100 nM. After an incubation period of 72 h, transfected cells were made quiescent and then treated with TGF-␤1. Subsequently, NADPH oxidase activity assay, Nox4, Poldip2, ␣-SMA, and Fn-EIIIA were examined as described above. Statistical analysis. All experiments were repeated at least three times. Experiments were analyzed by one-way ANOVA followed by Newman-Keuls multiple-comparison post hoc test performed by GraphPad Prism 5 software (GraphPad Prism, San Diego, CA). Significance was assigned at P ⱕ 0.05.

determine if Rac1 regulates myofibroblast activation, Rac1 GTP loading after stimulation with TGF-␤1 was examined at time intervals that we have previously reported to stimulate an early increase in Nox4 protein expression in kidney myofibroblasts (5). Short-term treatment of cells with TGF-␤1 had no effect on the level of active Rac1-GTP, as determined by a pulldown assay (Fig. 1, A and B). The Rac1 inhibitor NSC23766 substantially reduced active Rac1-GTP levels (Fig. 1B) but had no effect on TGF-␤1-induced increases in Nox4 protein (Fig. 1C) or NADPH oxidase activity as assayed by lucigeninenhanced chemiluminescence (Fig. 1D). *P ⬍ 0.05 vs. control. #P ⬍ 0.05 vs. TGF-ß1 according to ANOVA. As observed in a previous study (5), TGF-␤1 stimulated kidney fibroblasts to transition to a myofibroblast phenotype as assessed by increased expression of ␣-SMA and Fn-EIIIA. In resting fibroblasts, there was negligible expression of ␣-SMA and Fn-EIIIA as assessed by immunoblot analysis (Fig. 2, A

RESULTS

TGF-␤1-induced activation of kidney myofibroblast differentiation is not Rac1 dependent. Cellular signaling by several NADPH oxidase homologs requires the cytosolic subunit Rac1; however, an essential role for this small G protein in TGF-␤-induced Nox4 signaling of kidney myofibroblast differentiation is less certain (see the discussion above). To

A

B

TGF-β1 0

5

15

F161

30

60

min Cont.

Rac1-GTP

TGF-β1 TGF-β1+ NSC-23766 NSC-23766

Rac1-GTP Total Rac1

Total Rac1

1.6

Rac1-GTP/ Total Rac1

1.2

0.8

0.4

100 80

#

60

#

20

Cont.

TG

F-

β1 +N

Time in Minutes


Nox4 GAPDH

D

66 N SC -2 37 66

60

-2 37

30

SC

15

Fβ

5

C on tr ol

0

1

0

0.0

C

#

40

TG

Rac1-GTP/ Total Rac1

120

Fig. 1. Transforming growth factor (TGF)-␤1induced myofibroblast signaling via Nox4 is independent of Rac-GTP. A: TGF-␤1 had no effect on active Rac-GTP from 5 to 60 min, a time course known to stimulate myofibroblast differentiation (5) (see other figures). In additional experiments, cells were incubated with or without NSC-23766 (50 ␮M), an inhibitor of Rac, 2 h before the addition of TGF-␤1 or diluent. B–D: Rac-GTP loading (B), Nox4 protein (C), and NADPH oxidase activity (D) were examined after 30 min of incubation and compared with diluent controls. NSC-23766 reduced active Rac-GTP but had no effect on Nox4 expression (C) or NADPH oxidase activity (D). Cont, control. *P ⬍ 0.05 vs. control. #P ⬍ 0.05 vs. TGF-␤1 according to ANOVA.


F162


and B) and immunohistochemistry (Fig. 2, C and G). TGF-␤1 substantially increased both ␣-SMA and Fn-EIIIA (Fig. 2, A, B, D, and H). TGF-␤1 also induced myofibroblast transition characterized by the acquisition of ␣-SMA into well-defined stress fibers as well as Fn-EIIIA substrata expression viewed by immunofluorescence relative to controls. NSC-23766 did not reduce TGF-␤1-induced increased expression of ␣-SMA and Fn-EIIIA by immunoblot analysis or immunohistochemistry (Fig. 2, A, B, E, and I). These results indicate that Rac1 does not participate in TGF-␤1-induced Nox4-mediated kidney myofibroblast differentiation.

A

Cont.

TGF-␤1 rapidly stimulates RhoA and ROCK activation. Because RhoA GTPase is a known target of TGF-␤1 (20, 48), we then focused on a role for this G protein in signaling kidney myofibroblast differentiation. We observed that TGF-␤1 stimulated Rho GTP loading as early as 5 min after the addition of TGF-␤1 to kidney fibroblasts (Fig. 3A). TGF-␤1 continued increasing the level of RhoA-GTP up to 1 h, peaking at 30 min. Total RhoA remained unaffected by TGF-␤1. Similarly, protein expression of ROCK, an immediate effector of activated Rho-GTP, increased at 5 min after the addition of TGF-␤1, again peaking at 30 min (Fig. 3B). Additionally, TGF-␤1

B



Fn-EIIIA

α-SMA GAPDH

C

Cont.

GAPDH

D

E

F

H

I

J

α-SMA G

Fn-EIIIA Control

TGF-β1

TGF-β1 + NSC-23766

NSC-23766

Fig. 2. TGF-␤1-induced kidney myofibroblast differentiation is not regulated through Rac-1. Kidney fibroblasts were stimulated with TGF-␤1 or diluent with or without pretreatment with NSC-23766, as described in Fig. 1. Myofibroblast differentiation was examined 2 days later by detection of ␣-smooth muscle actin (␣-SMA) and alternatively spliced fibronectin variant (Fn-EIIIA) protein using Western blot analysis (A and B) and immunofluorescence microscopy (C–J). TGF-␤1 stimulated increased ␣-SMA protein incorporation in stress fibers by immunohistochemistry (D). Incubation of cells with NSC-23766 did not reduce TGF-␤1-induced increased expression of ␣-SMA or Fn-EIIIA, as shown in Western blots (A and B) or by immunohistochemistry (E and I). Similarly, ␣-SMA-positive stress fibers were not altered by NSC-23766 (E). *P ⬍ 0.05 vs. control according to ANOVA. AJP-Renal Physiol • doi:10.1152/ajprenal.00546.2013 • www.ajprenal.org

F163

KIDNEY MYOFIBROBLAST ACTIVATION VIA THE RhoA/ROCK/Poldip2/Nox4 PATHWAY TGF-β1

A 0

5

15

TGF-β1

B 30

60

min

0

RhoA-GTP

ROCK

Total Rho

GAPDH

RhoA-GTP/Total RhoA

9

5

15

TGF-β1

C 30

60

min

0

5

15

30

60

min

pMYPT1 Total MYPT1

*

8 7 6 5 4 3 2 1 0 0

5

15

30

60

Time in Minutes

Fig. 3. TGF-␤1 rapidly activates the RhoA/Rho kinase (ROCK) signaling pathway. A: TGF-␤1 stimulated myofibroblast signaling of active RhoA-GTP from 5 to 60 min. B and C: TGF-␤1 also stimulated the downstream effector ROCK (B) and phosphorylation of the ROCK substrate myosin phosphatase targeting protein-1 (MYPT-1; C), as detected by immunoblot analysis of lysates collected 5–30 min after the addition of TGF-␤1 or diluent. pMYPT-1, phosphorylated MYPT-1. *P ⬍ 0.05 vs. control according to ANOVA.

stimulated phosphorylation of MYPT-1, the immediate substrate for ROCK, used as an indicator of its activity, closely correlating with the activation of RhoA and ROCK (Fig. 3C). RhoA/ROCK mediates TGF-␤1-induced myofibroblast activation assessed by ␣-SMA and Fn-EIIIA expression. The above experiments indicated that the activation of the RhoA/ROCK pathway is an early event in kidney fibroblast activation by TGF-␤1. To further examine a role for this pathway in kidney myofibroblast differentiation, cells were exposed to siRhoA before the addition of TGF-␤1. The results showed that downregulation of RhoA protein with siRhoA inhibited RhoA expression relative to nontargeting siRNAs (Fig. 4D). siRhoA significantly reduced the expression of both ␣-SMA and Fn-EIIIA relative to controls as detected by immunoblot analysis (Fig. 4, A–C). Similarly, the effect of ROCK inhibitor Y-27632 on myofibroblast differentiation was ex-

A

B

amined (Fig. 5, A–F). Y-27632 inhibited TGF-␤1-stimulated increases in both ␣-SMA and Fn-EIIIA expression by Western blot analysis (Fig. 5, A–C), which was substantiated by immunofluorescence histochemistry (Fig. 5, D–F). These results indicate that RhoA/ROCK plays a role in myofibroblast activation by TGF-␤1. Role of RhoA/ROCK in ROS-mediated myofibroblast activation through NADPH oxidase. Our previous study (5) indicated that TGF-␤1-induced kidney myofibroblast differentiation is dependent on Nox4-derived ROS. To determine a role for the RhoA/ROCK pathway in TGF-␤1-induced Nox4/ROS stimulation of myofibroblast differentiation, the effects of siRhoA and Y-27632 on Nox4 protein expression and NADPH oxidase activity were investigated. As previously shown, TGF-␤1 stimulated a significant increase in Nox4 protein and NADPH oxidase activity (Fig. 6, A–F). TGF-␤1-induced increases in

C

α-SMA Fn-EIIIA GAPDH ntRNA

ntRNA siRho A siRhoA +TGF-β1 +TGF-β1

D RhoA GAPDH ntRNA

siRhoA

Fig. 4. TGF-␤1-induced myofibroblast differentiation requires RhoA signaling. Cells were exposed to small interfering (si)RNA to RhoA (siRhoA) or nontargeting (nt)RNA for 72 h. TGF-␤1 or diluent was then added, and lysates were collected 24 h later to examine markers of myofibroblast differentiation by immunoblot analysis. TGF-␤1-enhanced ␣-SMA and Fn-EIIIA protein expression was reduced by siRhoA relative to ntRNA (A–C). D: verification of the effectiveness of siRhoA to inhibit RhoA protein relative to ntRNA. *P ⬍ 0.05 vs. ntRNA control. #P ⬍ 0.05 vs. ntRNA ⫹ TGF-␤1 according to ANOVA. AJP-Renal Physiol • doi:10.1152/ajprenal.00546.2013 • www.ajprenal.org

F164


A

B

C

α-SMA Fn-EIIIA GAPDH Cont

TGF-β1 TGF-β1 Y27632 +Y27632

D

E

F

G

H

I

J

F

Control

TGF-β1

TGF-β1 + Y27632

Y-27632

Fig. 5. TGF-␤1-induced myofibroblast differentiation is blocked by ROCK inhibition. A–C: addition of Y-27632 (10 ␮M), an inhibitor of ROCK, 1 h before TGF-␤1 also blocked myofibroblast differentiation, as illustrated by the reduced expression of ␣-SMA and Fn-EIIIA relative to controls. D, E, H, and I: TGF-␤1 stimulated an increase in ␣-SMA protein expression incorporation in stress fibers (E) as well as Fn-EIIIA protein (I) relative to controls (D and H), as illustrated by immunofluorescence histochemistry. F and J: The ROCK inhibitor Y-27632 blocked ␣-SMA (F) and Fn-EIIIA (J) protein expression. F and G: diluent had no effect on the expression of ␣-SMA or Fn-EIIIA. *P ⬍ 0.05 vs. control. #P ⬍ 0.05 vs. TGF-␤1 according to ANOVA.

Nox4 protein expression were significantly reduced by siRhoA (Fig. 6, A and B) as well as by Y-27632 (Fig. 6, D and E). Similarly, TGF-␤1-enhanced NADPH oxidase activity was abrogated by siRhoA (Fig. 6C) as well as Y-27632 (Fig. 6F). TGF-␤1 stimulated a significant increase in intracellular ROS, as revealed by increased DCF fluorescence detected by confocal microscopy (Fig. 6G). Y-27632 nearly completely inhibited DCF fluorescence (Fig. 6G). Diphenyleneiodonium, an inhibitor of NADPH oxidase and other flavin-containing oxidases, had no effect on RhoA-GTP loading by TGF-␤1 (Fig. 6, H and I) but inhibited NADPH oxidase activity (Fig. 6J), indicating that NADPH oxidase-derived ROS are downstream of RhoA. Overall, the above data show that RhoA/ROCK regulates TGF-␤1-induced myofibroblast activation through Nox4-dependent ROS generation. Role of Poldip2/Nox4 in TGF-␤1-induced myofibroblast activation. A role for Poldip2, a known regulator of Nox4, in kidney myofibroblast differentiation was examined by transduction of fibroblasts with AdPoldip2 and compared with control adenovirus without construct. Likewise, the effect of Nox4 on Poldip2 expression and myofibroblast differentiation was examined using Nox4 adenovirus. The results showed that overexpression of Poldip2 protein examined 2 days after in-

fection enhanced basal expression of Nox4 (Fig. 7, A–C) and enhanced NADPH oxidase activity (Fig. 7D) relative to controls. Similarly, overexpression of Poldip2 protein via AdPoldip2 enhanced the basal expression of myofibroblast differentiation markers ␣-SMA and Fn-EIIIA (Fig. 7, A, E, and F). Conversely, we found that overexpression of Nox4 protein via AdNox4 enhanced basal expression of Poldip2 (Fig. 7, G–I) and enhanced NADPH oxidase activity (Fig. 7J) relative to controls. Similarly, overexpression of Nox4 protein enhanced the basal expression of myofibroblast differentiation markers ␣-SMA and Fn-EIIIA (Fig. 7, G, K, and L). We also found that treatment of cells with TGF-␤1 increased Poldip2 and Nox4 protein expression (Fig. 8, A–C) and NADPH oxidase activity (Fig. 8D) relative to controls as early as 30 min after stimulation. Knockdown of Poldip2 using siPoldip2 inhibited TGF␤1-mediated increases in Nox4 protein and NADPH oxidase activity as well as downstream markers of myofibroblast activation (␣-SMA and Fn-EIIIA) relative to nontargeting controls (Fig. 8, A, E, and F). RhoA regulates Poldip2/Nox4 in TGF-␤1-induced myofibroblast activation. To explore a role for the RhoA/ROCK pathway in the regulation of Poldip2/Nox4 by TGF-␤1, cells were transfected with siRhoA or treated with Y-27632 before the



A

B

C

E

F

F165

Nox4 Rho-GTP Total Rho GAPDH ntRNA ntRNA siRhoA siRhoA + TGF-β1 +TGF-β1

D Nox4 GAPDH

Cont

TGF-β1 TGF-β1 Y27632 +Y27632

G DCF Control

I

100 80

*

*

60 40 20

D PI

+D

FTG

TG

on tr o C

PI

β1

0

l

Control TGF-β1 TGF-β1 DPI + DPI

J *

β1

Total Rho

Y-27632

F-

RhoA-GTP

TGF-β1 + Y27632

120

RhoA-GTP/ Total RhoA

H

TGF-β1

Fig. 6. The RhoA/ROCK pathway signals Nox4-derived ROS. A and C: kidney fibroblasts were treated with siRhoA or ntRNA for 72 h followed by stimulation TGF-␤1 or diluent, and Nox4 expression (A) and NADPH oxidase activity (C) were then examined 30 min later. A–C: siRhoA inhibited the TGF-␤1-induced increased expression of Nox4 protein (A and B) and NADPH oxidase activity (C). D–F: similarly, cells treated with the ROCK inhibitor Y-27632 before the addition of TGF-␤1 blocked the increased expression of Nox4 protein (D and E) and NADPH oxidase activity (F) relative to diluent controls. A–F: inhibitors alone had no effect on basal levels of Nox4 protein or NADPH oxidase activity. G: ROS generation, detected by 2=,7=-dichlorodihydrofluorescein (DCF) fluorescence, was also inhibited by Y-27632 within 30 min after incubation with TGF-␤1. H–J: inhibition of Nox4 using diphenyleneiodonium (DPI) had no effect on TGF-␤1-induced activation of RhoA-GTP (H and I) but reduced NADPH oxidase activity (J), indicating that RhoA is upstream of Nox4. *P ⬍ 0.05 vs. control. #P ⬍ 0.05 vs. TGF-␤1 according to ANOVA.

addition of TGF-␤1. The results showed that inhibition of either RhoA or ROCK resulted in a reduction in TGF-␤1induced increases in Poldip2 protein (Fig. 9, A and B). Overexpression of Poldip2 in kidney fibroblasts infected with AdPoldip2 had no effect on RhoA-GTP loading or expression of ROCK or phosphorylated MYPT-1 protein relative to cells infected with adenovirus with green fluorescent protein, indicating that Poldip2 is downstream of RhoA/ROCK signaling

(Fig. 9, C–G). Also, inhibition of ROCK using Y-27632 had no effect on AdPoldip2-induced overexpression of ␣-SMA or Fn-EIIIA proteins (Fig. 9, H–J). These results indicate the specificity of Poldip2 in myofibroblast differentiation by bypassing the Rho/ROCK pathway. Collectively, the results above indicate that the RhoA/ROCK pathway plays a significant role in regulating myofibroblast activation through modulation of Poldip2 and Nox4.


F166


A

B

Myc

C

Poldip2 Nox4 α-SMA Fn-EIIIA GAPDH AdGFP AdPoldip2

D

F

*

8 6 4 2

2

0 ld Ad

Po

Ad

G

ip

FP

NADPH-dependent superoxide generation (RLU/min/mg protein)

I

0

2

4 ox dN A

dG A

No Ad

GF Ad

FP

x4

0


x4 No

G

1

Ad

*

3

2

1

0

x4

2

*

4

FP

4

L

3

G

P

*

No

GF

6

Ratio FN/GAPDH Protein

K

8

Ad

Ad

AdNox4

P

J

NADPH-dependent superoxide generation (RLU/min/mg protein)

AdGFP

0

FP

0

GAPDH

1

No

4

2

Ad

Fn-EIIIA

8

* 3

Ad

α-SMA

Ratio Poldip2/GAPDH Protein

Poldip2

12

4

Ratio α -SMA/GAPDH Protein

Nox4

*

16

Ad

H

x4

G

Ratio Nox4/GAPDH Protein

Fig. 7. Polymerase (DNA-directed) ␦-interacting protein 2 (Poldip2) regulates myofibroblast differentiation via Nox4. A–F: kidney fibroblasts were infected with AdPoldip2 or adenovirus with green fluorescent protein (AdGFP) over a 48-h period, and the relative effects of overexpression of Poldip2 on Nox4 protein (A–C), NADPH oxidase activity (D), and ␣-SMA and Fn-EIIIA (A, E, and F) expression were examined. Increased expression of Poldip2, as determined by Myc and Poldip protein by immunoblot analysis, resulted in enhanced expression of Poldip2 and Nox4 protein and NADPH oxidase activity as well as the myofibroblast differentiation markers ␣-SMA and Fn-EIIIA relative to control. G–L: infection of kidney fibroblasts with AdNox4 significantly increased Nox4 and Poldip2 protein (G–I), NADPH oxidase activity (J), and the myofibroblast differentiation markers ␣-SMA and Fn-EIIIA (G, K, and L) relative to control (AdGFP). *P ⬍ 0.05 vs. AdGFP vs. control.

E


A

B

C

E

F

F167

Nox4 Poldip2 α-SMA Fn-EIIIA GAPDH ntRNA ntRNA siPoldip2 siPoldip2 +TGF-β +TGF-β

D

Fig. 8. TGF-␤1-induced myofibroblast differentiation is regulated via signaling via Poldip2 and Nox4. A–D: kidney fibroblasts were treated with siRNA to Poldip2 (siPoldip2) or ntRNA for 72 h followed by TGF-␤1 or diluent, and Poldip2 and Nox4 expression (A–C) and NADPH oxidase activity (D) were then examined 30 min later. siPoldip2 inhibited TGF-␤1-induced increases in Poldip2 and Nox4 protein expression as well as NADPH oxidase activity relative to control (ntRNA). siPoldip2 also inhibited TGF-␤1-induced myofibroblast differentiation, as illustrated by reductions in ␣-SMA (A and E) and Fn-EIIIA (A and F) relative to ntRNA. *P ⬍ 0.05 vs. ntRNA control. #P ⬍ 0.05 vs. ntRNA ⫹ TGF-␤1 according to ANOVA.

DISCUSSION

TGF-␤1 signaling of myofibroblast differentiation is known to occur through two pathways: Smad and ERK (1, 42). Details of these transduction pathways have proved to be complex, may vary in different cell types, and remain incompletely understood (3, 28). Our previous study (5) showed that TGF␤1-induced kidney myofibroblast differentiation, characterized by ␣-SMA and Fn-EIIIA expression, is regulated by Nox4derived ROS via a signaling cascade involving Smad3, Nox4generated ROS, and ERK1/2. The present study shows that Rac1 is not involved in downstream Nox4 regulation of kidney myofibroblast differentiation. Rather, the small GTPase RhoA/ ROCK pathway regulates Nox4-dependent ROS generation and subsequent downstream cellular events in TGF-␤1-induced kidney myofibroblast activation. TGF-␤1 rapidly activates RhoA and ROCK, which, in turn, phosphorylates MYPT-1, an indicator of ROCK activity. siRhoA and the ROCK inhibitor Y-27632 both abolished TGF-␤1-induced increases in Nox4 protein and NADPH oxidase activity as well as expression markers of myofibroblast differentiation (␣-SMA and FnEIIIA). These results and those of our previous study collectively indicate that TGF-␤1 induces kidney myofibroblast differentiation as assessed by ␣-SMA expression and Fn-EIIIA synthesis via a redox signaling cascade involving the TGF-␤1 receptor (TGF-␤1R)1/RhoA/ROCK/Nox4/Poldip2 pathway (Fig. 10). As in our previous studies (3, 5), activation of these signaling events occurred within minutes, consistent with

translational mechanisms in Nox4 signaling. A novel role for Poldip2 in signaling of kidney myofibroblast differentiation was indicated by the finding that Poldip2 enhanced Nox4 protein and NADPH oxidase activity and subsequent ␣-SMA and Fn-EIIIA expression. Conversely, siPoldip2 inhibited TGF-␤1-induced Nox4 upregulation, ROS generation, and myofibroblast differentiation. This demonstrates that Poldip2 is required for TGF-␤1-induced Nox4-dependent ROS production and subsequent myofibroblast activation and acts as an upstream regulator of Nox4 expression and ROS release. Here, we provide the first evidence that Poldip2 is a downstream effector of RhoA in that siRhoA and Y27643 inhibited Poldip2 protein expression and myofibroblast differentiation. It is well established that TGF-␤1 binds TGF-␤R type II (TGF-␤RII) and subsequently phosphorylates a type I receptor (TGF-␤RI) subunit, forming a heterodimeric complex (1, 42) that signals through a canonical pathway involving the Smad family of transcriptional activators in myofibroblasts. TGF␤R1 phosphorylates Smad2/3, which subsequently complexes with Smad4 and translocates to the nucleus, where the dimer binds to the promoter region of the ␣-SMA gene (3, 28). ROS play a significant role in the progression of fibrotic disease (23, 32, 52). ROS have been identified as intermediates in TGF-␤ signaling of mesangial cells or fibroblasts during fibrosis or progressive disease, which has received considerable recent interest (3). Indeed, ROS in the form of superoxide and hydrogen peroxide are directly linked to transmodulation of



A

B

Poldip2

Poldip2

GAPDH

GAPDH ntRNA ntRNA siRhoA siRhoA + TGF-b1 +TGF-b1

C

Myc

D

E

TGF-β1 TGF-β1 Y27632 +Y27632

F

G

RhoA-GTP Total Rho ROCK GAPDH pMYPT1

α-SMA Fn

100

#

*

*

#

60

#

40 20 0

dG A

fibroblasts to ␣-SMA myofibroblasts (67) and are important mediators of fibrogenic cellular responses including hypertrophy, migration, proliferation, apoptosis, and regulation of the extracellular matrix (5, 17, 23, 56, 60). NADPH oxidases are important sources of ROS involved in both normal physiological functions and oxidative stress (3). Of the NADPH oxidases of the Nox family, the Nox4 homolog is now recognized as a principal source of ROS generation and subsequent redoxdependent pathological processes in progressive fibrotic disease (3). TGF-␤1 is a major regulator of Nox4 protein expression and Nox4-dependent ROS production in a number of cell types, including smooth muscle cells, endothelial cells, hepatocytes, and fibroblasts (3). The results of the present study in kidney fibroblasts also agree with our previous studies and other published studies indicating a close relationship between ROS generated by NADPH oxidase and ␣-SMA expression

#

80

A dG

GAPDH

J 120

120 100

#

*

#

*

80 60

# #

40 20 0 A dG dG FP FP A + dP Yol 2 76 di p2 32 +Y -2 76 32 A dP ol di p2

I

FP FP A +Y dP -2 ol 76 di p2 32 +Y -2 76 32 A dP ol di p2

H

Fn-EIIIA/GAPDH Protein

Total pMYPT1

a-SMA/GAPDH Protein

Fig. 9. RhoA/ROCK regulate TGF-␤1-induced Poldip2 protein expression. A: kidney fibroblasts were treated with siRhoA or ntRNA for 72 h followed by TGF-␤1 or diluent, and lysates were then collected 30 min later for immunoblot analysis of Poldip2 protein. siRhoA inhibited TGF-␤1-induced increases in Poldip2 protein relative to ntRNA control. B: in additional experiments, kidney fibroblasts were treated with the ROCK inhibitor Y-27632 before the addition of TGF-␤1 or diluent, and Poldip2 protein expression was determined by immunoblot analysis. Inhibition of ROCK reduced TGF␤1-induced increases in Poldip2 protein expression relative to controls. C–G: overexpression of Poldip2 in kidney fibroblasts infected with AdPoldip2 had no effect on RhoA-GTP loading or expression of ROCK and pMYPT1 relative to cells exposed to AdGFP, indicating that Poldip2 is downstream of RhoA/ROCK signaling. H–J: inhibition of ROCK using Y-27632 had no effect on AdPoldip2-induced overexpression of ␣-SMA (H and I) or Fn-EIIIA (H and J), showing the specificity of Poldip2 in myofibroblast differentiation. *P ⬍ 0.05 vs. control. #P ⬍ 0.05 vs. TGF-␤1 according to ANOVA.

Cont

A

F168

and a myofibroblast phenotype in adventitial fibroblast and vascular smooth muscle cells (3, 5, 9, 10, 55). We (5) have previously shown that kidney myofibroblasts express NADPH oxidase homologs Nox2 and Nox4, with the latter appearing to be the most important in the generation of ROS in response to TGF-␤1 and downstream signaling of ERK1/2 and ␣-SMA and Fn-EIIIA expression. The mechanisms by which Nox enzymes are regulated in cardiovascular or renal cells and how they generate ROS are not fully understood. A number of regulatory subunits of Nox enzymes have been identified. Importantly, p22phox is required as a docking and maturation subunit for the homologs Nox1, Nox2, and Nox4 (3, 7). Activation mechanisms for Nox1 are similar to those of Nox2 but involve complex formation with regulatory cytosolic subunits Nox organizer 1, Nox activator 1, and the small G protein Rac1 (3, 7, 35, 36). The small GTPases



Fig. 10. Proposed signaling cascade involving RhoA/ROCK/Poldip2/Nox4 in TGF-␤1-induced kidney myofibroblast activation.

act as molecular switches, and it is well established that the Nox family of NADPH oxidases is principally regulated through a Rac-dependent pathway (8, 18, 29). Nox1 and Nox2 are tightly regulated by Rac1 through its binding to phox and Nox organizer 1 or Nox activator 1 subunits in Nox2 and Nox1 assembly. A role for Rac1 in Nox4 activation is uncertain. Although it has been suggested that Rac1 is implicated in the control of Nox4 function in mesangial and endothelial cells (8, 18), several studies in heterologous systems have shown that Nox4 does not require the recruitment of regulatory cytoplasmic subunits and Rac1 to be active. Nox4-mediated ROS generation is undisturbed in Rac1 knockdown (41) or chemical inhibition (4), supporting the notion that Nox4 is Rac1 independent and suggests alternate transduction pathways. Unlike mesangial cells, in which Rac promotes TGF-␤1-stimulated collagen type 1 (30) and Rac1 appears to be involved in Nox4 signaling (18), kidney myofibroblast do not use this GTPase in TGF-␤1 signaling of Nox4 activation of ␣-SMA and Fn-EIIIA. TGF-␤1 did not stimulate Rac1 activity, and inhibition of Rac1 had no effect on TGF-␤1-induced Nox4-dependent ROS production or ␣-SMA and Fn-EIIIA expression in kidney myofibroblasts. Rather, our data indicate that TGF-␤1/Nox4 signaling of myofibroblast differentiation implicates RhoA GTPase and ROCK. Of particular interest are experiments identifying the RhoA/ROCK system as an important signaling system regulating the actin cytoskeleton and stress fiber formation (39). The RhoA/ROCK system plays an important role in regulating

F169

the myofibroblast phenotype in mesenchymal progenitor cells, pericytes, and fibroblasts (12, 21, 34, 58). However, studies showing an association of ROS signaling and the RhoA/ROCK pathway in vitro are limited (29, 39, 64, 65). Studies in endothelial cells have indicated that Rac1 is necessary for downstream activation of RhoA and subsequent Nox4 activation and cell proliferation, whereas Rac1-dependent Nox2 activity was responsible for cytoskeletal rearrangement via JNK (63, 64). In vascular smooth muscle cells, RhoA was found to be downstream of Nox4 and associated with focal contacts and cell migration (39). Our experiments in kidney fibroblasts showed that RhoA is upstream of Nox4, as in endothelial cells, but similar to smooth muscle cells, in that TGF-␤1-induced kidney fibroblast differentiation to a myofibroblast phenotype is associated with the formation of ␣-SMA incorporated in stress fibers. The present study placing RhoA/ ROCK upstream of Nox oxidase is consistent with reports showing that the ROCK inhibitor fasudil significantly reduced angiotensin II-induced Nox4 upregulation and ROS production in the vascular endothelium (27) as well as diabetes-mediated increases in Nox4 mRNA expression in the renal cortex (16). An important recent discovery in smooth muscle cells shows that Poldip2 associates with Nox4 and stabilizes focal adhesions and stress fiber formation through RhoA (39) and is involved in the maintenance of a constitutively expressed smooth muscle phenotype. To date, our experiments with renal fibroblasts are the only other studies in which a Poldip2/Nox4 interaction has been identified and show regulation of the Nox4/Poldip2 interaction in response to TGF-␤1. Interestingly, our experiments showed that TGF-␤1 induces a reverse order of signaling events in kidney myofibroblasts compared with smooth muscle cells, where Poldip2 and Nox4 are downstream of RhoA. A RhoA/Poldip2/Nox4 sequence of transduction may indicate a novel outside-in pathway for ROS signaling at the level of the substratum. The differences between vascular smooth muscle cells and kidney myofibroblasts may reflect different states of stress fiber and matrix expression in the two cell types. Smooth muscle cells constitutively express ␣-SMA, whereas kidney fibroblast differentiation into myofibroblasts appears to be regulated by TGF-␤1 via RhoA/ROCK/Poldip2/ Nox4 interactions. Poldip2 has been shown to associate with Nox4 and control ROS production by Nox4 (39, 57). Here, we show that Poldip2 controls Nox4 expression and NADPH oxidase activity. Moreover, our data suggest that TGF-␤1mediated upregulation of Nox4 protein and NADPH oxidase activity depends on the induction of Poldip2 expression by TGF-␤1. A TGF-␤1-dependent increase in Poldip2 most likely translates to an increase in Nox4 expression since Poldip2 controls the expression of the oxidase. It is also conceivable that the increased Poldip2 levels promoted by TGF-␤1 are also a means to enhance Nox4-dependent ROS production. Importantly, all these events take place downstream of RhoA/ROCK activation by TGF-␤1. Also, we show, for the first time, that Nox4 can regulate Poldip2 expression, suggesting a functional association or positive feedback loop in the activation of these two proteins. The above experiments indicate that RhoA/ROCK-dependent regulation of Poldip2/Nox4 is involved in TGF-␤1-induced kidney myofibroblast activation in vitro. A number of studies have also indicated that RhoA/ROCK may be a key initiator in oxidative stress during disease where ROCK inhib-


F170


itors blocked the progression of cardiac hypertrophy, liver fibrosis, diabetic nephropathy, renal fibrosis, and glomerulosclerosis (16, 31, 43– 45). Of interest are the observations that oxidative stress or Nox1 and Nox4 are inhibited by ROCK inhibitors in several of these models (16, 45). Furthermore, myofibroblasts during renal fibrosis originate, at least in part, from a perivascular adventitium (2, 13, 38, 62). This observation is consistent with the known role of NADPH oxidasegenerated ROS in adventitial fibroblast activation in aortic perivascular disease (22, 49). Thus, a role for this RhoA/ ROCK/Poldip2/Nox4 pathway in interstitial myofibroblast encroachment in vivo during renal fibrosis is plausible but awaits further investigation. ACKNOWLEDGMENTS The authors thank Fredyne Springer and Christine Spencer for technical contributions and Dr. Karen Block for the rabbit anti-Nox4 antibody. GRANTS This work was supported by grants from the Veterans Administration Merit Review Program, National Institutes of Health (NIH) Grant RO1-DK-080106, and the NIH George O’Brien Kidney Center and Morphology Core. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: N.M., M.P., Y.G., and J.L.B. conception and design of research; N.M. and M.P. performed experiments; N.M., M.P., K.K.G., Y.G., and J.L.B. analyzed data; N.M., M.P., K.K.G., Y.G., and J.L.B. interpreted results of experiments; N.M., M.P., and J.L.B. prepared figures; N.M., M.P., and J.L.B. drafted manuscript; N.M., M.P., K.K.G., Y.G., and J.L.B. approved final version of manuscript; K.K.G., Y.G., and J.L.B. edited and revised manuscript. REFERENCES 1. Attisano L, Wrana JL. Signal transduction by the TGF-␤ superfamily. Science 296: 1646 –1647, 2002. 2. Barnes JL, Glass WF. Renal interstitial fibrosis: a critical evaluation of the origin of myofibroblasts. In: Experimental Models for Renal Diseases: Pathogenesis and Diagnosis, edited by Herrera GA. Basel: Karger, 2011, p. 73–93. 3. Barnes JL, Gorin Y. Myofibroblast differentiation during fibrosis: role of NAD(P)H oxidases. Kidney Int 79: 944 –956, 2011. 4. Basuroy S, Bhattacharya S, Leffler CW, Parfenova H. Nox4 NADPH oxidase mediates oxidative stress and apoptosis caused by TNF-␣ in cerebral vascular endothelial cells. Am J Physiol Cell Physiol 296: C422– C432, 2009. 5. Bondi CD, Manickam N, Lee DY, Block K, Gorin Y, Abboud HE, Barnes JL. NAD(P)H oxidase mediates TGF-␤1-induced activation of kidney myofibroblasts. J Am Soc Nephrol 21: 93–102, 2010. 6. Borsi L, Castellani P, Risso AM, Leprini A, Zardi L. Transforming growth factor-␤ regulates the splicing pattern of fibronectin messenger RNA precursor. FEBS Lett 261: 175–178, 1990. 7. Brandes RP, Weissmann N, Schroder K. NADPH oxidases in cardiovascular disease. Free Radic Biol Med 49: 687–706, 2010. 8. Chai D, Wang B, Shen L, Pu J, Zhang XK, He B. RXR agonists inhibit high-glucose-induced oxidative stress by repressing PKC activity in human endothelial cells. Free Radic Biol Med 44: 1334 –1347, 2008. 9. Clempus RE, Sorescu D, Dikalova AE, Pounkova L, Jo P, Sorescu GP, Lassegue B, Griendling KK. Nox4 is required for maintenance of the differentiated vascular smooth muscle cell phenotype. Arterioscler Thromb Vasc Biol 27: 42–48, 2007. 10. Cucoranu I, Clempus R, Dikalova A, Phelan PJ, Ariyan S, Dikalov S, Sorescu D. NAD(P)H oxidase 4 mediates transforming growth factor-␤1induced differentiation of cardiac fibroblasts into myofibroblasts. Circ Res 97: 900 –907, 2005.

11. Desmouliere A, Geinoz A, Gabbiani F, Gabbiani G. Transforming growth factor-␤1 induces ␣-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 122: 103–111, 1993. 12. Diah S, Zhang GX, Nagai Y, Zhang W, Gang L, Kimura S, Hamid MR, Tamiya T, Nishiyama A, Hitomi H. Aldosterone induces myofibroblastic transdifferentiation and collagen gene expression through the Rho-kinase dependent signaling pathway in rat mesangial cells. Exp Cell Res 314: 3654 –3662, 2008. 13. Faulkner JL, Szcykalski LM, Springer F, Barnes JL. Origin of interstitial fibroblasts in an accelerated model of angiotensin II (Ang II)induced renal fibrosis. Am J Pathol 167: 1193–1205, 2005. 14. Gaedeke J, Peters H, Noble NA, Border WA. Angiotensin II, TGF-␤, and renal fibrosis. Contrib Nephrol 135: 153–160, 2001. 15. Ghosh C, Abboud HE. Tyrosine phosphorylation-dependent PI3 kinase/ Akt signal transduction regulates TGF-␤-induced fibronectin expression in mesangial cells. Cell Signal 16: 31–41, 2004. 16. Gojo A, Utsunomiya K, Taniguchi K, Yokota T, Ishizawa S, Kanazawa Y, Kurata H, Tajima N. The Rho-kinase inhibitor, fasudil, attenuates diabetic nephropathy in streptozotocin-induced diabetic rats. Eur J Pharmacol 568: 242–247, 2007. 17. Gorin Y, Block K, Hernandez J, Bhandari B, Wagner B, Barnes JL, Abboud HE. Nox4 NAD(P)H oxidase mediates hypertrophy and fibronectin expression in the diabetic kidney. J Biol Chem 280: 39616 – 39626, 2005. 18. Gorin Y, Ricono JM, Kim NH, Bhandari B, Choudhury GG, Abboud HE. Nox4 mediates angiotensin II-induced activation of Akt/protein kinase B in mesangial cells. Am J Physiol Renal Physiol 285: F219 –F229, 2003. 19. Griendling KK, Ushio-Fukai M. Reactive oxygen species as mediators of angiotensin II signaling. Regul Pept 91: 21–27, 2000. 20. Harvey KA, Paranavitana CN, Zaloga GP, Siddiqui RA. Diverse signaling pathways regulate fibroblast differentiation and transformation through Rho kinase activation. J Cell Physiol 211: 353–363, 2007. 21. Haudek SB, Gupta D, Dewald O, Schwartz RJ, Wei L, Trial J, Entman ML. Rho kinase-1 mediates cardiac fibrosis by regulating fibroblast precursor cell differentiation. Cardiovasc Res 83: 511–518, 2009. 22. Haurani MJ, Cifuentes ME, Shepard AD, Pagano PJ. Nox4 oxidase overexpression specifically decreases endogenous Nox4 mRNA and inhibits angiotensin II-induced adventitial myofibroblast migration. Hypertension 52: 143–149, 2008. 23. Haurani MJ, Pagano PJ. Adventitial fibroblast reactive oxygen species as autacrine and paracrine mediators of remodeling: bellwether for vascular disease? Cardiovasc Res 75: 679 –689, 2008. 24. Hayashida T, Decaestecker M, Schnaper HW. Cross-talk between ERK MAP kinase and Smad signaling pathways enhances TGF-␤-dependent responses in human mesangial cells. FASEB J 17: 1576 –1578, 2003. 25. Hayashida T, Wu MH, Pierce A, Poncelet AC, Varga J, Schnaper HW. MAP-kinase activity necessary for TGF␤1-stimulated mesangial cell type I collagen expression requires adhesion-dependent phosphorylation of FAK tyrosine 397. J Cell Sci 120: 4230 –4240, 2007. 26. Heasman SJ, Ridley AJ. Mamalian Rho GTPases: new insights into their functions from in vivo studies. Nat Rev Mol Cell Biol 9: 690 –701, 2008. 27. Higashi M, Shimokawa H, Hattori T, Hiroki J, Mukai Y, Morikawa K, Ichiki T, Takahashi S, Takeshita A. Long-term inhibition of Rhokinase suppresses angiotensin II-induced cardiovascular hypertrophy in rats in vivo: effect on endothelial NAD(P)H oxidase system. Circ Res 93: 767–775, 2003. 28. Hinz B. Formation and function of the myofibroblast during tissue repair. J Invest Dermatol 127: 526 –537, 2007. 29. Hordijk PL. Regulation of NADPH oxidases. The role of Rac proteins. Circ Res 98: 453–462, 2006. 30. Hubchak SC, Sparks EE, Hayashida T, Schnaper HW. Rac1 promotes TGF-␤-stimulated mesangial cell type I collagen expression through a PI3K/Akt-dependent mechanism. Am J Physiol Renal Physiol 297: F1316 –F1323, 2009. 31. Kanda T, Wakino S, Hayashi K, Homma K, Ozawa Y, Saruta T. Effect of fasudil on Rho-kinase and nephropathy in subtotally nephrectomized spontaneously hypertensive rats. Kidney Int 64: 2009 –2019, 2003. 32. Kanwar YS, Sun L, Xie P, Liu FY, Chen S. A glimpse of various pathogenetic mechanisms of diabetic nephropathy. Annu Rev Pathol 6: 395–423, 2011.


KIDNEY MYOFIBROBLAST ACTIVATION VIA THE RhoA/ROCK/Poldip2/Nox4 PATHWAY 33. Kawamura S, Miyamoto S, Brown JH. Initiation and transduction of stretch-induced RhoA and Rac1 activation through caveolae: cytoskeletal regulation of ERK translocation. J Biol Chem 278: 31111–31117, 2003. 34. Kolyada AY, Riley KN, Herman IM. Rho GTPase signaling modulates cell shape and contractile phenotype in an isoactin-specific manner. Am J Physiol Cell Physiol 285: C1116 –C1121, 2003. 35. Lambeth JD, Kawahara T, Diebold B. Regulation of Nox and Duox enzymatic activity and expression. Free Radic Biol Med 43: 319 –331, 2007. 36. Lassegue B, San Martin A, Griendling KK. Biochemistry, physiology, and pathophysiology of NADPH oxidases in the cardiovascular system. Circ Res 110: 1364 –1390, 2012. 37. Liang CC, Park AY, Guan JL. In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat Protoc 2: 329 –333, 2007. 38. Lin SL, Kisseleva T, Brenner DA, Duffield JS. Pericytes and perivascular fibroblasts are the primary source of collagen-producing cells in obstructive fibrosis of the kidney. Am J Pathol 173: 1617–1627, 2008. 39. Lyle AN, Deshpande NN, Taniyama Y, Seidel-Rogol B, Pounkova L, Du P, Papaharalambus C, Lassegue B, Griendling KK. Poldip2, a novel regulator of Nox4 and cytoskeletal integrity in vascular smooth muscle cells. Circ Res 105: 249 –259, 2009. 40. Mahadev K, Motoshima H, Wu X, Ruddy JM, Arnold RS, Cheng G, Lambeth JD, Goldstein BJ. The NAD(P)H oxidase homolog NOX4 modulates insulin-stimulated generation of H2O2 and plays an integral role in insulin signal transduction. Mol Cell Biol 24: 1844 –1854, 2004. 41. Martyn KD, Frederick LM, von LK, Dinauer MC, Knaus UG. Functional analysis of Nox4 reveals unique characteristics compared to other NADPH oxidases. Cell Signal 18: 69 –82, 2006. 42. Massague J, Gomis RR. The logic of TGF-␤ signaling. FEBS Lett 580: 2811–2820, 2006. 43. Moriyama T, Kawada N, Nagatoya K, Takeji M, Horio M, Ando A, Imai E, Hori M. Fluvastatin suppresses oxidative stress and fibrosis in the interstitium of mouse kidneys with unilateral ureteral obstruction. Kidney Int 59: 2095–2103, 2001. 44. Nagatoya K, Moriyama T, Kawada N, Takeji M, Oseto S, Murozono T, Ando A, Imai E, Hori M. Y-27632 prevents tubulointerstitial fibrosis in mouse kidneys with unilateral ureteral obstruction. Kidney Int 61: 1684 –1695, 2002. 45. Nishikimi T, Akimoto K, Wang X, Mori Y, Tadokoro K, Ishikawa Y, Shimokawa H, Ono H, Matsuoka H. Fasudil, a Rho-kinase inhibitor, attenuates glomerulosclerosis in Dahl salt-sensitive rats. J Hypertens 22: 1787–1796, 2004. 46. Okuda S, Languino LR, Ruoslahti E, Border WA. Elevated expression of transforming growth factor-beta and proteoglycan production in experimental glomerulonephritis. Possible role in expansion of the mesangial extracellular matrix. J Clin Invest 86: 453–462, 1990. 47. Parizi M, Howard EW, Tomasek JJ. Regulation of LPA-promoted myofibroblast contraction: role of Rho, myosin light chain kinase, and myosin light chain phosphatase. Exp Cell Res 254: 210 –220, 2000. 48. Peng F, Zhang B, Wu D, Ingram AJ, Gao B, Krepinski JC. TGF-␤induced RhoA activation and fibronectin production in mesangial cells requires caveolae. Am J Physiol Renal Physiol 295: F153–F164, 2008. 49. Rey FE, Pagano PJ. The reactive adventitia. Fibroblast oxidase in vascular function. Arterioscler Thromb Vasc Biol 22: 1962–1971, 2002.

F171

50. Ridley AJ. Rho family proteins: coordinating cell responses. Trends Cell Biol 11: 471–477, 2001. 51. Riento K, Ridley AJ. ROCKs: multifunctional kinases in cell behaviour. Nat Rev Mol Cell Biol 4: 446 –456, 2003. 52. Sachse A, Wolf G. Angiotensin II-induced reactive oxygen species and the kidney. J Am Soc Nephrol 18: 2439 –2446, 2007. 53. San Martín A, Griendling KK. Redox control of vascular smooth muscle migration. Antioxid Redox Signal 12: 625–640, 2010. 54. Serini G, Bochaton-Piallat ML, Ropraz P, Geinoz A, Borsi L, Zardi L, Gabbiani G. The fibronectin domain ED-A is crucial for myofibroblastic phenotype induction by transforming growth factor-␤1. J Cell Biol 142: 873–842, 1998. 55. Shen WL, Gao PJ, Che ZQ, Ji KD, Yin M, Yan C, Berk BC, Zhu DL. NAD(P)H oxidase-derived reactive oxygen species regulate angiotensin-II induced adventitial fibroblast phenotypic differentiation. Biochem Biophys Res Commun 339: 337–343, 2006. 56. Sundaresan M, Yu ZX, Ferrans VJ, Irani K, Finkel T. Requirement for generation of H2O2 for platelet-derived growth factor signal transduction. Science 270: 296 –299, 1995. 57. Sutliff RL, Amanso AM, Parastatidis I, Dikalova AE, Hansen L, Datla SR, Long JS, El-Ali AM, Gleason JG Jr, Taylor WR, Hart CM, Griendling KK, Lassègue B. Polymerase delta interacting protein 2 sustains vascular structure and function. Arterioscler Thromb Vasc Biol 33: 2154 –2161, 2013. 58. Vardouli L, Vasilaki E, Papadimitriou E, Kardassis D, Stournaras C. A novel mechanism of TGF␤-induced actin reorganization mediated by Smad proteins and Rho GTPases. FEBS J 275: 4074 –4087, 2008. 59. Venkatesan B, Mahimainathan L, Das F, Ghosh-Choudhury N, Ghosh Choudhury G. Downregulation of catalase by reactive oxygen species via PI 3 kinase/Akt signaling in mesangial cells. J Cell Physiol 211: 457–467, 2007. 60. Weber DS, Seshiah PN, Taniyama Y, Griendling KK. Src-dependent migration of vascular smooth muscle cells by PDGF is reactive oxygen species dependent. Circulation 106: S260, 1995. 61. Welsh CF, Roovers K, Villanueva J, Liu Y, Schwartz MA, Assoian RK. Timing of cyclin D1 expression within G1 phase is controlled by Rho. Nat Cell Biol 3: 950 –957, 2001. 62. Wiggins R, Goyal M, Merritt S, Killen PD. Vascular adventitial cell expression of collagen I messenger ribonucleic acid in anti-glomerular basement membrane antibody-induced crescentic nephritis in the rabbit. A cellular source for interstitial collagen synthesis in inflammatory renal disease. Lab Invest 68: 557–565, 1993. 63. Wu RF, Ma Z, Liu Z, Terada LS. Nox4-derived H2O2 mediates endoplasmic reticulum signaling through local Ras activation. Mol Cell Biol 30: 3553–3568, 2010. 64. Wu RF, Ma Z, Myers DP, Terada LS. HIV-1 Tat activates dual Nox pathways leading to independent activation of ERK and JNK MAP kinases. J Biol Chem 282: 37412–37419, 2007. 65. Wu RF, Xu YC, Ma Z, Nwariaku FE, Sarosi GA Jr, Terada LS. Subcellular targeting of oxidants during endothelial cell migration. J Cell Biol 171: 893–904, 2005. 66. Yue J, Mulder KM. Requirement of Ras/MAPK pathway activation by transforming growth factor ␤ for transforming growth factor ␤1 production in a Smad-dependent pathway. J Biol Chem 275: 30765–30773, 2000. 67. Zalewski A, Shi Y. Vascular myofibroblasts. Lessons from coronary repair and remodeling. Arterioscler Thromb Vasc Biol 17: 417–422, 1997.


Am J Physiol Renal Physiol 307: F172–F182, 2014. First published June 4, 2014; doi:10.1152/ajprenal.00215.2014.

Macroscopic electrical propagation in the guinea pig urinary bladder F. T. Hammad,1 B. Stephen,2 L. Lubbad,1 J. F. B. Morrison,2 and W. J. Lammers2 1

Department of Surgery, United Arab Emirates University, Al Ain, United Arab Emirates; and 2Department of Physiology, College of Medicine and Health Sciences, United Arab Emirates University, Al Ain, United Arab Emirates

Submitted 23 April 2014; accepted in final form 30 May 2014

Hammad FT, Stephen B, Lubbad L, Morrison JF, Lammers WJ. Macroscopic electrical propagation in the guinea pig urinary bladder. Am J Physiol Renal Physiol 307: F172–F182, 2014. First published June 4, 2014; doi:10.1152/ajprenal.00215.2014.—There is little knowledge about macroscopic electrical propagation in the wall of the urinary bladder. Recording simultaneously from a large number of extracellular electrodes is one technology that could be used to study the patterns of macroscopic electrical propagations. The urinary bladders from 14 guinea pigs were isolated and placed in an organ bath. A 16 ⫻ 4-electrode array was positioned at various sites on the serosal bladder surface, and recordings were performed at different intravesical volumes. In four experiments, carbachol (CCH; 10⫺6 M), nifedipine (10 mM), or tetrodotoxin (TTX; 10⫺6 M) was added to the superfusing fluid. After the experiments, the extracellular signals were analyzed and propagation maps were constructed. Electrical waves were detected at all sites on the bladder surface and propagated for a limited distance before terminating spontaneously. The majority of waves (⬎90%) propagated in the axial direction (i.e., from dome to base or vice versa). An increase in vesicle volume significantly decreased the conduction velocity (from 4.9 ⫾ 1.5 to 2.7 ⫾ 0.7 cm/s; P ⬍ 0.05). CCH increased, nifedipine decreased ,while TTX had little effect on electrical activities. In addition, a new electrical phenomenon, termed a “patch,” was discovered whereby a simultaneous electrical deflection was detected across an area of the bladder surface. Two types of electrical activities were detected on the bladder surface: 1) electrical waves propagating preferentially in the axial direction and 2) electrical patches. The propagating electrical waves could form the basis for local spontaneous contractions in the bladder during the filling phase. urinary bladder; electrical impulse; electrical propagation

activities of the urinary bladder have been investigated in many studies in the past decades. The introduction of microelectrodes for intracellular recordings has made it possible to study in detail the electrical events in normal and diseased detrusor cells (38, 46). Other technologies, such as optical mapping, have recently been introduced to study the propagation of action potentials and calcium waves in the whole bladder (30). Another mapping technology, multielectrode mapping with a large number of extracellular electrodes, has been used to analyze the propagation of electrical impulses in other smooth muscle organs, such as the stomach (33), the myometrium (34), the renal pelvis (32), and the ureters (21, 22) under normal and pathological conditions. In the current study, using the same technology, we have recorded the electrical activities from several parts of the guinea pig bladder at different bladder volumes and analyzed their patterns of propagation. In all experiments, we could detect electrical waves originating from different sites on the bladder

THE ELECTRICAL AND MECHANICAL

Address for reprint requests and other correspondence: Wim J. E. P. Lammers, Dept. of Physiology, College of Medicine and Health Sciences, POBox 17666, Al Ain, United Arab Emirates (e-mail: [email protected]). F172

and propagating predominantly in the axial direction before terminating spontaneously. These waves could be the basis for the reported micromotions or contraction waves that occur spontaneously in the urinary bladder during the filling phase (11, 12, 47). It has been proposed that the intramural plexuses of the bladder may be involved in the generation of the reported micromotions (14), and in the present experiments we used tetrodotoxin to block nerve-mediated responses in an attempt to ascertain whether the intrinsic nerves were involved in the observed electrical activities. METHODS

Studies were performed in 14 male guinea pigs (weight 780 ⫾ 119 g; ⬃8 –12 mo old). The experimental protocol was approved by the Animal Research Ethics Committee, College of Medicine and Health Sciences, United Arab Emirates University. Following an overnight fast during which the guinea pigs were given water ad libitum, the animals were anesthetized with ether inhalation. After the animals reached a sufficient depth of anesthesia (as tested by the absence of withdrawal reflexes), through a midline abdominal incision the urinary bladder and urethra with surrounding structures (seminal vesicles, vas deferens, prostate) were removed en bloc and placed immediately in a 400-ml organ bath, which contained a modified Tyrode solution (composition in mM: 130 NaCl, 4.5 KCl, 2.2 CaCl2, 0.6 MgCl2, 24.2 NaHCO3, 1.2 NaH2PO4, and 11 glucose). This was followed by animal euthanasia by ether overdosing. The isolated bladder was superfused in the organ bath at a rate of 100 ml/min with the modified Tyrode solution, which was saturated with carbogen (95% O2 and 5% CO2) and kept at a constant pH (7.35 ⫾ 0.05) and temperature (37 ⫾ 0.5oC). The urethra was cannulated and connected to a pressure transducer (AD Instruments), and the intravesical pressure was recorded on a PowerLab Chart (Fig. 1A). Through a Y-connector, the catheter was also connected to a 10-ml syringe, which was used to fill the bladder with different volumes to study the effect of bladder volume on electrical activity. Electrophysiological recordings were performed with a 64-electrode probe (16 ⫻ 4) array; the interelectrode distance along each row was 0.35 mm, while the distance between the rows was ⬃0.9 mm (Fig. 1B). The electrodes were made of 0.30-mm-diameter Tefloncoated silver wires that had been chlorided before the experiments. The electrode array was carefully positioned, using a micromanipulator, sequentially on the serosal surface of the dome, the base, and across the peritoneal reflection on the anterior surface of the bladder, under constant visual supervision, thereby avoiding any significant compression of the bladder wall. In all experiments, recordings were performed from the anterior surface of the bladder, while in three experiments recordings were also performed from the dorsal surface. Unipolar electrical recordings were performed with a large silver plate located in the tissue bath as the indifferent pole. The 64 recording electrodes were connected through shielded wires to 64 AC preamplifiers, where the signals were amplified (4,000 times), filtered (2– 400 Hz), and digitized (1-kHz sampling rate) before being stored as previously described (32) (Fig. 1C). After the experiments, the signals were analyzed with custom-made software (Smoothmap.exe; www.smoothmap.org).


http://www.ajprenal.org

ELECTRICAL ACTIVITIES IN THE GUINEA PIG BLADDER

A Organ bath and preparation: to amplifiers

pressure transducer

64-electrodes

bladder

B 64-electrode probe:

F173

the distance, time difference, and angle from the first to the last activation time in every individual wave (Fig. 2C3). To estimate the degree of electrical activity in the bladder, a column of 16 recording electrodes was chosen that showed maximal electrical activity in a particular recording. An example of this procedure is shown in Fig. 9B. All 16 electrograms in that column were examined to determine whether electrical activity was present for every second during a period of 1 min. The results are presented in percentages, whereby 0% showed a total lack of electrical activity in all 16 electrograms and 100% is the situation in which electrical activity was detected all the time throughout an entire minute. The effect of carbachol (CCH), tetrodotoxin (TTX), or nifedipine on bladder activity was studied in four animals each. In these experiments, the bladder was filled with 4 ml Tyrode and CCH (10⫺6 M), TTX (10⫺6 M), or nifedipine (10 mM) was added to the superfusing Tyrode. Recordings were performed during control and during drug application, from the three different regions of the bladder, as previously described. All data are presented as means and standard deviations. Student’s t-test was used in this study to determine statistical significance. A P value of ⬍0.05 was considered statistically significant. RESULTS

C Mapping System:

hard disk Amplifiers & Filters

Electrodes

AD-converter

PC-Monitor

PC-computer

PC-Monitor

Fig. 1. A: experimental set-up with a diagram of the organ bath, isolated bladder, and catheter connections. B: close-up of the tip of the electrode array. Four rows of 16 Teflon-coated silver wires each were glued together in this 4 ⫻16 array. C: diagram of the multiple electrode recording system. See text for further descriptions.

For the off-line analysis, signals were digitally filtered (20-point moving average; low-pass filter of ⬃40 Hz) to remove the 50-Hz noise and then displayed on-screen in sets of 16 electrograms (Fig. 2A). The local activation time of a wave was identified by the moment of maximum negative slope and marked with a cursor (21, 32), whereby the reference time was determined by the timing of the first detected wave (t ⫽ 0.0 s at electrode 3 in column c in Fig. 2B). After all the waves had been so marked, their activation times were displayed in the format of a grid of the original recording array of electrodes (Fig. 2C1). In case of a low-quality signal or if no deflections were visible, those sites were represented by empty circles in the maps. To visualize the sequence of propagation, isochrones were drawn manually around areas activated in time intervals of 20 ms and false colored (Fig. 2C2). From the data, an estimate was also made of the direction of propagation and the velocity by calculating

In the recordings from the bladder surface, electrical impulses occurred in single, isolated spikes or in groups of closely related spikes (bursts). These impulses were detected from all the recorded areas including the dome, the base, and the area of the peritoneal reflection. For instance, Fig. 3 shows a typical waveform at the beginning of a burst as recorded from the dome. Two propagating waveforms can be followed as bior triphasic deflections of relatively high amplitudes. Wave a originated from outside the mapped area and stopped propagating ⬃0.5 mm below the upper edge of the map (Fig. 3B). Wave b appeared at that upper border 50 ms later and propagated in the opposite direction before it also stopped propagating spontaneously in the lower part of the map. Propagation could also be detected in the base of the bladder and in the neighborhood of the peritoneal reflection. As shown in Fig. 4, at the initiation of a burst, several wave sequences could be seen and their maps revealed that the impulse was initiated twice, at the same location in the mapped area, and that the two waves propagated toward the peritoneal reflection. A phenomenon that was also observed in all recordings is the disintegration of propagation in time. For example, in the electrograms in Fig. 4B, following the first two propagating waves, the third and fourth waves became less clear, with broader, longer lasting deflections and lower amplitudes. In addition to the demonstration of propagating waves, another electrical phenomenon was discovered in the guinea pig bladder, an example of which is shown in Fig. 5. In this recording of the dome, a wave was seen propagating from the peritoneal reflection towards the apex of the dome. The corresponding propagation map shows a homogenously propagating wave that took ⬃180 ms to reach the upper edge of the map. A few seconds later, in the same area, another deflection was detected (Fig. 5, right). The amplitude of these deflections was largest in the middle of the map, at electrodes 8 and 9, and gradually decreased away from this area until it became undetectable at electrodes 1 and 16. Moreover, we could not detect a time delay between the deflections, as indicated by the vertical dotted line, suggesting that these signals had occurred at the same time. In Fig. 5B, all 16 electrograms from both


F174


Peritoneal Reflection

Dome

A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

0.05 mV

B

col b

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

col c

C1

a

b

c

3 13 28 43 59 65 74 83 102 117 130 151 154

0 19 40 53 65 77 90 105 117 126 139

d

C2

a

1.0 mm


col d

0 19 40 53 65 77 90 105 117 126 139

3 13 28 43 59 65 74 83 102 117 130 151 154

0.05 mV

b

c

Dome

Fig. 2. Analysis of propagating waves. A: 16 electrograms recorded from column b of the electrode array, located on the bladder dome, for a period of 10 s. Several bouts of electrical activities are visible, one of which is selected for further analysis (rectangle). B: electrograms recorded at all 4 electrode columns in that selected period. A negative deflection is visible in several electrograms in columns b and c but not in columns a and d. C1: local activation times are plotted onto a grid of the original electrode array. C2: local times have been left out, and the isochronal areas have been false colored together with additional symbols. Block symbols (㛳) have been added to indicate areas where propagation was blocked. C3: method by which the direction and velocity of propagation of this wave were determined.

1 second

col a

d

C3

a

200 msec

b

0

c

d

0 40

Dome 90

80

180

o

0 270

120

160

o

o

o

Base

154

Direction: 258 o Velocity: 2.67 cm/s

events were superimposed. In the left panel, the succession of waveforms, lasting ⬃180 ms, is evident. In the right panel, all intrinsic deflections occurred simultaneously. The corresponding maps in Fig. 5C show the difference between a propagating wave (left) and this new phenomenon, which we have termed an electrical “patch” (right panel). As indicated in METHODS, we calculated the velocities of the propagating waves, and the results are plotted in Fig. 6. The average velocity of the propagating waves was 3.6 ⫾ 1.6 cm/s, and there were no significant differences in velocities at the three different bladder locations (dome: 3.5 ⫾ 1.6 cm/s, peritoneal reflection: 4.6 ⫾ 1.7 cm/s, base: 3.1 ⫾ 1.4 cm/s). However, when the values were plotted against the bladder

volumes, there was a clear and significant reduction in velocity at higher volumes from 4.9 ⫾ 1.5 ml at 2-ml to 2.7 ⫾ 0.7 cm/s at 8-ml bladder volume (P ⬍ 0.05). The direction of propagations was also analyzed, and the results are shown in Fig. 7. There was a very strong prevalence for the waves to propagate toward the dome or toward the base of the bladder in a longitudinal direction while waves propagating in the transverse direction were rare (13/142; 9.2%). An example of such a lateral propagating wave is shown in Fig. 8. In Fig. 8A1, a large wave was seen propagating from left to right. The electrograms recorded in the second column (Fig. 8A2, column b) show more or less simultaneous activity as the wave front reaches the electrode column at about the same


F175


Fig. 3. Recording from the dome (A) showing 2 waves propagating in opposite directions (B). Recordings were made from column c. Both waves originated from outside but stopped propagating within the mapped area. Note the block symbols (㛳) indicating where the propagating waves had stopped spontaneously. Bladder volume ⫽ 4 ml.

time. However, electrograms recorded in the transverse direction, as shown in rows A, B, and C in Fig. 8A3, demonstrate clear propagation from one electrode to the next. To validate this preferential propagation in the longitudinal direction, in seven recordings (2 experiments), the electrode array was rotated 90°, with its long axis oriented in the transverse direction. In this condition, the majority (84%) of waves still propagated either toward the dome or toward the base. An overall summary of the recordings is presented in Table 1. From a total of 53 recordings, performed in 14 experiments, 142 propagating waves and 80 patches were analyzed. Propagating waves occurred more frequently than patches at all three bladder locations. Furthermore, there was no prevalence for

waves or patches to occur more frequently in any particular area. Effect of bladder volume on electrical activity. Distension of the bladder always induced a significant increase in bursts of electrical activity. As shown in Fig. 9A, this was the case in the three regions investigated: dome, peritoneal reflection, and base. This relationship was quantified at these three sites and plotted in Fig. 9C. Effect of CCH on electrical activity. In four experiments, after control recordings, CCH (10⫺6 M) was added to the superfusing Tyrode. Recordings were performed from the three different regions of the bladder. A total of 29 waves and 9 patches were analyzed. In general, electrical propagation in the


F176


Fig. 4. Recordings from the base of the bladder and at the beginning of a burst of activity (bladder volume ⫽ 2 ml). A: 1-min recording from 1 electrode (electrode 9 in B) that displays repetitive bursts of electrical activity. B: 16 electrograms selected from the beginning of a burst, as indicated by the rectangle in A. The displayed electrograms were recorded from column c (as indicated in C). C: propagation maps of the 2 waves indicated in B. The 2 waves originated spontaneously from the middle of the area (stars) and propagated toward the peritoneal reflection. Propagation in the lateral direction was blocked.

presence of CCH was very similar to those previously described in the absence of CCH. Similarly, the majority of waves (⬎90%) propagated in the longitudinal direction as was the case during control. However, propagation velocity was lower in CCH (2.78 ⫾ 1.19 cm/s in CCH vs. 3.66 ⫾ 1.56 cm/s in control, P ⬍ 0.05). Finally, using the counting method described in METHODS and illustrated in Fig. 9B, electrical activity during CCH administration was shown to be significantly increased (from 41 ⫾ 18 to 84 ⫾ 15%, P ⬍ 0.001). Effect of nifedipine on electrical activity. In another four experiments, nifedipine (10 mM) was added to the superfusing Tyrode. A total of 22 waves and 16 patches were analyzed. In

all four preparations, nifedipine significantly decreased electrical activity from 25 ⫾ 11% (average ⫾ SD) to 7 ⫾ 7% (P ⬍ 0.001). Similarly, velocity was also decreased (from 4.67 ⫾ 1.71 to 3.81 ⫾ 1.37 cm/s, P ⬍ 0.01). Effect of TTX on electrical activity. The addition of TTX (10⫺6 M) to the superfusing Tyrode did not have much effect on the electrical activities in the bladder. Both propagating waves (n ⫽ 31) as well as patches (n ⫽ 13) occurred in the detrusor wall, and the degree of electrical activity was similar to that recorded during control (from 20 ⫾ 12% in control to 29 ⫾ 22% in TTX; not significant), although velocity was slightly decreased (from 4.04 ⫾ 1.53 to 3.26 ⫾ 1.35 cm/s, P ⬍ 0.05).


F177


Fig. 5. Two types of electrical activities in the bladder (volume ⫽ 4 ml). A, left: wave propagating across the dome; right, deflections and the area excited during the activation of a “patch.” B: superimpositon of all 16 electrograms for both wave and patch. C, left: isochrone colors indicate time in steps of 20 ms; right, isochrone colors indicate voltage in steps of 0.01 mV. See text for a further description.

DISCUSSION

This study presents, to the best of our knowledge, the first demonstration of detailed macroscopic electrical propagation in the urinary bladder. In this study, we showed that 1) electrical waves originate spontaneously from any site on the bladder wall and propagate for a finite distance until spontaneous termination; 2) the vast majority of these waves propagated either toward the dome or toward the base; 3) the patterns of propagation were similar at the dome, both ventral and dorsal surfaces, across the peritoneal reflection and at the base; 4) the increase in bladder

volume significantly decreased conduction velocity; 5) CCH (10⫺6 M), TTX (10⫺6 M), and nifedipine (10 mM) all significantly reduced conduction velocity; and 6) CCH significantly increased whereas nifedipine significantly decreased overall spontaneous activity. Finally, we report an electrical phenomenon, new to the bladder electrophysiology literature, termed an electrical patch, wherein simultaneous electrical activity was observed in different areas of the bladder wall. Several studies have characterized the electrical activities of the urinary bladder in several species including humans, usu-


F178


vs. Bladder Volume

7.0

7.0

6.0

6.0

Velocity (cm/s)

Fig. 6. Left: plot of velocities of propagating waves at 3 different locations in the bladder (n ⫽ 69). Right: decreasing velocities at increasing bladder volumes (n ⫽ 65). *Significant difference from velocities at 2 ml (P ⬍ 0.05).

Velocity (cm/s)

vs.

5.0 4.0 3.0 2.0 1.0

5.0 4.0

*

3.0 2.0 1.0 0.0

0.0

Dome


ally with microelectrodes (10, 39, 40, 46) but occasionally also with other techniques such as the double sucrose gap (17), optical scanning (30), and extracellular electrodes. Boyce (5) was one of the first to explore the possibilities of recording bladder electromyograms in humans but concentrated on the low frequencies (⬍1 Hz), which was later shown by Brunsting (8) to be caused by motion artifacts. Later, Craggs and Stephenson (9), Kinder et al. (31), Scheepe et al. (41, 42), and Ballaro et al. (3) recorded extracellular signals using suction or silver electrodes and described the waveforms of the detrusor muscle with a bi- or triphasic morphology that was very similar to what we recorded in this study. Obviously, motion artifacts can also be recorded with these types of electrodes, as can also be seen in the tracings in Fig. 9, but the frequency component of these motion-induced signals is much lower than that of the electrical signals. There is evidence that electrical propagation in the bladder is mostly limited to individual bundles and that large-scale propagation of waves, such as described in this study, are not that common. Tomita et al. (44) described the low safety factor in propagation when an action potential approaches the branching of functional bundles, which could lead to the “spread of excitation to be blocked.” Bramich and Brading (7) injected current into one cell while recording from neighboring cells and showed that most cells did not record any deflection at all,

Base

2 ml

4 ml

6 ml

8 ml

and, if cells did register a deflection, then these were most likely located in the axial direction. Fry et al. (18) studied electrical coupling in strips of guinea pig detrusor and compared them to similar strips from the cardiac ventricle. Interestingly, although both tissues showed similar space constants, the values of the intracellular resistance and that of the cellular membranes were much higher in the detrusor muscle cells, suggesting far fewer intercellular electrical couplings in the bladder compared with the heart. Kanai et al. (30), using optical scanning of propagating depolarizations and calcium waves, showed in adult rat bladders disorganized propagations from many sites of origin, whereas in the neonatal bladders, propagations were much more organized and originated mostly from the suburothelium region (30). From a macroscopic point of view, in the adult guinea pig bladder, this was also our experience. In our recordings, the vast majority of electrical activities consisted of low-amplitude, high-frequency spikes, often barely above the noise level (41). Neighboring electrodes showed similar rapid deflections, but most were not in synchrony with the deflections at adjacent electrodes, suggesting that local microscopic propagation is very limited, even within the dimension of the tip of a single recording electrode; i.e., 0.30 mm, roughly equivalent to 5–50 cells (23).

Fig. 7. Analysis of direction of propagation of 142 individual waves. Histogram shows the number of waves propagating in a particular direction grouped in 10° bins. The y-axis plots the number of waves in a particular bin. The diagram reveals a strong preference for wave propagation to occur in the longitudinal direction, either toward the dome or, in the opposite direction, toward the base, and seldom in the transverse direction.


Copyright of American Journal of Physiology: Renal Physiology is the property of American Physiological Society and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Industry and rural health: part of the problem or part of the solution?

Are you part of the problem?

Are we part of the problem?

Crista supinatoris fractures of the proximal part of the ulna.

Peri-implantitis. Part 1: Scope of the problem.

Reverse shoulder arthroplasty for the treatment of three-part and four-part proximal humeral fractures in the elderly.

Medication adherence part one: understanding and assessing the problem.

Surgical exposure of the lesser trochanter and the medial proximal part of the femur.

The emigration of doctors: a problem for the developing and the developed countries. Part II.

The phosphatidylinositol synthase of proximal tubule cells.

The proximal tubule and albuminuria: really!

James Hillman: the unmaking of a psychologist. Part two: The problem of the puer.

Emigration of doctors: a problem for the developing and the developed countries. Part I.

The lineshape of the electronic spectrum of the green fluorescent protein chromophore, part II: solution phase.

Surgical management of metastatic disease of the proximal part of the femur.

Receptor-Mediated Endocytosis in the Proximal Tubule.

DGCR6 at the proximal part of the DiGeorge critical region is involved in conotruncal heart defects.

Facing the challenge of sustainable bioenergy production: Could halophytes be part of the solution?

Urea secretion by the straight segment of the proximal tubule.

Micronutrient deficiencies in the elderly - could ready meals be part of the solution?

Results of the initial administrations of the NBME comprehensive Part I and Part II examinations.

Psychology and mind-body segregation: Are we part of the problem?

Diabetic foot ulcers: Part II. Management.

The Local Authorities' Part.