Renal expression of monocyte chemotactic protein-1 and epidermal growth factor in children with obstructive hydronephrosis

Renal Expression of Monocyte Chemotactic Protein-1 and Epidermal Growth Factor in Children With Obstructive Hydronephrosis By Fabio Bartoli, Loreto Gesualdo, Guglielmo Paradies, Eustacchio Caldarulo, Barbara Infante, Giuseppe Grandaliano, Raffaella Monno, Samuele Leggio, Franca Salzillo, Francesco Paolo Schena, and Antonio Leggio Bari, Italy

Background/Purpose: The authors studied the potential role of ureteropelvic junction obstruction (UPJ-O) in causing progressive renal damage in children through the renal expression of epidermal growth factor (EGF) and monocyte chemotactic protein-1 (MCP-1) mRNA.

of UTI had a more severe reduction of the renal thickness of the affected kidney compared with those without UTI. MCP-1 expression was higher and EGF more reduced in children with a history of UTI.

Methods: Renal tissues were harvested from 11 children with UPJ-O and from 10 normal kidneys to study the renal expression of EGF and MCP-1 detected by means of in situ hybridization. Five of the patients were found to have a history of urinary tract infection (UTI).

Conclusions: Our results suggest a potential role of EGF and MCP-1 in the pathogenesis of renal damage and growth failure in UPJ-O, especially in children with UTI. These important functional changes begin early in life, possibly during fetal life. J Pediatr Surg 35:569-572. Copyright 娀 2000 by W.B. Saunders Company.

Results: Children with UPJ-O had marked reduction of EGF gene expression when compared with controls. Interstitial expression of MCP-1 mRNA was present in all UPJ-O cases. Both EGF and MCP-1 expression did not correlate with age, with differential renal function, and with renal thickness measured through MAG3 renal scan. Children with a history

O

BSTRUCTIVE UROPATHY is a common disorder in pediatric surgery. Treatment of ureteropelvic junction obstruction (UPJ-O) ranges from early surgical repair1 to long-term conservative management.2 Currently, there is a great interest in the investigation of the potential role of UPJ-O as a cause of progressive renal damage in children. It has been reported that obstructive nephropathy is characterized by glomurulosclerosis, interstitial inflammation, tubular atrophy, and monocyte infiltration,3 even though the basic mechanism underlying these histological changes remains unknown. The kidney is a major site of production of the epidermal growth factor (EGF),4 which seems to have several important functions in the modulation of renal growth, glomerular hemodinamics, renal metabolism, tubular transport, and eicosanoid synthesis.5 Furthermore, EGF appears to govern renal cell proliferation during fetal life6 and to maintain the epithelial turnover of the ductal system.5 Some investigators have reported a decreased expression of EGF mRNA in several renal diseases7 and in experimental hydronephrosis.8 Monocyte chemotactic protein-1 (MCP-1) is a powerful and specific chemotactic and activating factor for monocytes. An important role has been proposed for MCP-1 in the inflammatory process in the kidney.9 In addition, an increased MCP-1 gene expression recently has been reported in an experimental model of hydronephrosis.10 For these reasons we studied Journal of Pediatric Surgery, Vol 35, No 4 (April), 2000: pp 569-572

INDEX WORDS: Uretero pelvic junction obstruction, urinary tract infection, monocyte chemotactic protein-1, epidermal growth factor.

both EGF and MCP-1 mRNA expression in the renal biopsy sections of children operated on for UPJ-O and the correlation of this expression with the main clinical and histological features. MATERIALS AND METHODS Over a 12-month period, we studied 11 children (mean age, 2 years; range, 1 month to 14 years) with unilateral UPJ-O who had undergone pyeloplasty and renal biopsy. UPJ-O was diagnosed by ultrasound scan, cystography, intravenous pyelography, and MAG3 renography. In all patients, MAG3 renal scan was performed previous intravenous hydration and without a bladder catheter because it would modify the physiological pyelo-vesical emptying. After having injected a bolus of MAG3 technetium Tc 99 (dose, 50 µCi/kg), a basic observation was done for the first 20 minutes. The patients who still had an abnormal urinary drainage defined as the inability to eliminate at least 50% of marked substance present into the pelvis (T50), underwent furosemide stimulation, and then the T50 was monitored for additional 30 minutes. Pyelography and renography were performed under diuretic stimulation (furosemide, 0.5 mg/kg bolus intravenous) to discriminate between obstructive and nonobstructive pelvic dilatation (NOPD). Five patients had a prenatal diagnosis of hydronephrosis. In all cases, vesico-ureteral reflux and megaureter were ruled out. All patients had normal blood

From the Cattedra di Chirurgia Pediatrica, Dipartimento di Biomedicina dell’Eta` Evolutiva, Universita` di Bari, Italy. Address reprint requests to Dott. Fabio Bartoli, Cattedra di Chirurgia Pediatrica, Dipartimento di Biomedicina dell’Eta` Evolutiva, Piazza G. Cesare 11, I-70124, Bari, Italy. Copyright 娀 2000 by W.B. Saunders Company 0022-3468/00/3504-0007$03.00/0 569

570

BARTOLI ET AL

urea nitrogen and creatinine levels. In none of them were clinical signs or markers of active infection at the time of surgery detected. The transverse diameter of the renal parenchima measured by MAG3 scan in both kidneys was considered as indirect expression of renal parenchimal thickness (RT). The differences in RT between the affected and nonaffected side were reported as percentage reduction of renal thickness (prRT). Four children had NOPD and were operated on, 1 because of recurrent urinary tract infection (UTI), 1 for severe abdominal pain, and 2 because differential function (DF) less than 40%. Five of the children had a history of UTI. Table 1 summarizes the main clinical features of the patients enrolled in the study. With previous informed consent, all children underwent Anderson-Hynes pyeloplasty and intraoperative renal biopsies of affected kidney. Ten control renal biopsy specimens from apparently normal renal tissue taken from nephrectomized children were obtained. Nephrectomy was indicated in 6 because of Wilms’ tumor, 3 because of infiltrating neuroblastoma, and 1 because of traumatic renal injury.

In Situ Hybridization Renal tissue was included in embending medium compound (OCT; Tissue-Tek, Elkhart, IN), snap-frozen in liquid nitrogen, and stored in the same liquid until used. Frozen sections (6 µm thick) were collected onto polylysine-coated slides, briefly dried at 80°C, fixed in 4% paraformaldehyde for 20 minutes, washed twice in phosphate-buffered saline (PBS), and dehydrated in graded ethanol. For the preparation of the MCP-1 RNA probe, a 200 base pair (bp) cDNA template was used, which had been obtained by reverse transcriptase polymerase chain reaction (RT-PCR) of total RNA extracted from cultured human mesangial cells using 2 specific primers (sense primer: 5⬘ TTCTGTGCCTGCTGCTCA 3⬘, antisense primer: 5⬘ CTCACTCACAAGTTCAGA 3⬘) deduced from the human full-length cDNA sequence.11 The resulted PCR product was sequenced and subcloned into the PCRII vector (InVitrogen Corp, San Diego, CA). For the preparation of specific EGF RNA probes a fragment of EGF of 1000 bp obtained from the complete cDNA was subcloned into the plasmid pGEM1 (Promega Biotech, Florence, Italy). After linearization of the plasmids with the appropriate restriction enzymes, T7 or SP6 RNApolymerase (Boehringer Mannheim, FRG), respectively, were used to obtain runoff transcripts of either antisense or sense strands. Transcription, labeling of RNA probes, prehybridization, hybridization, removal of specifically nonbound probe by RNase A digestion, and further washing procedures were performed for both sense and antisense MCP-1 RNA probes as previously described.12 Autoradiography was performed by dipping the dehydrated slides into Ilford G5 nuclear emulsion (Ilford, Mobberley Cheshire, United

Table 1. Clinical Features of Patients Enrolled in Study Age (d)

Differential Function (%)

prRT (%)

1,440 4,680 180 18 270 2,200 120 60 60 240 25

48 3 47 73 53 39 28 47 3 5 49

5 ⫺49 ⫺18 ⫺8 ⫺32 20 10 ⫺24 ⫺40 ⫺48 ⫺17

UTI

EGF (AU)

Tubular Damage (AU)

MCP-1/CD 68 (AU)

No No Yes No Yes No No No Yes Yes Yes

3.5 3 2.5 2 2 1.5 1.5 1.5 1 0.5 0

⫺ ⫹/⫺ ⫹ ⫹ ⫹/⫺ ⫹ ⫹⫹ ⫹/⫺ ⫹⫹ ⫹⫹ ⫹⫹⫹

1.5/2 0.5/0.5 0.5/1 0.5/1 1.5/3 3/3 1/1 0.5/1 1.5*/3 3*/3 1.5*/3

*Extension of MCP-1 gene expression also at tubular and glomerular levels.

Kingdom). The exposed slides were developed using Kodak D19 developer (Kodak, Hemel Hampstead, United Kingdom), stained counter in H&E and finally mounted.

Immunohistochemistry The immunohistochemical detection of monocytes was performed on 4-µm-thick frozen kidney sections using chromatographically purified mouse anti-CD68 monoclonal antibody specific for human monocytes (Dako, Milan, Italy). Immobilized mouse antibody was detected by the immunoalkaline phosphatase (APAAP) method as previously described.12 Alkaline phosphatase was developed with New Fuchsin (Sigma, Milan, Italy). Negative controls were performed by omitting either the primary or secondary antibodies and using nonimmune mouse antiserum as first layer.

Histological Study Biopsy specimens also were fixed in Bouin’s fluid, dehydrated in alcohol, embedded in paraffin, and cut in 2-µm-thick sections. The sections were stained with H&E, and the histological score was compared with in situ hybridization studies. MCP-1 and EGF gene expression, the extent of monocyte infiltration, and the degree of tubular damage (tubular atrophy and dilatation) were assessed semiquantitatively by 2 independent blind observers. The degree of tubular damage, the degree of CD68 infiltrate, and the amount of gene expression for EGF and MCP-1 were expressed as arbitrary units (AU).

Statistical Analysis Data were expressed as mean ⫾ SD. The differences between groups were tested with unpaired t test (Student’s test), linear regression, and analysis of variance.

RESULTS

In control biopsy specimens, a high expression of EGF was observed (Fig 1A), whereas specific hybridization for MCP-1 mRNA was undetectable (Fig 2A). In UPJ-O kidney biopsy specimens EGF expression was significantly downregulated and completely absent in atrophic and dilated tubules (Fig 1B). However, pronounced increase in MCP-1 gene expression was observed. Specific hybridization for this chemokine was located mainly on interstitial and glomerular parietal epithelial cells (Fig 2B). In Fig 3 are reported the semiquantitative values of EGF and MCP-1, which show a statistically significant difference for both signals between the control group and the UPJ-O group. Interestingly, EGF gene expression was inversely correlated with the degree of tubular damage observed under the light microscope (Table 1). To know whether MCP-1 increased expression was associated with monocyte influx into the interstitial space, the extent of CD68-positive cell infiltration was evaluated (Fig 4). As expected, the number of monocytes was directly and significantly correlated with MCP-1 mRNA expression (r ⫽ 0.86; P ⫽ .05; Table 1). Neither EGF nor MCP-1 expression correlated with the age, the differential renal function, and the renal thickness as measured by MAG3 renal scan at the time of diagnosis (Table 1). In addition, MCP-1 gene expression was higher in

RENAL EXPRESSION OF MCP-1 AND EGF IN UPJ-O

571

Fig 1. In situ hybridization for EGF. Dark-field photomicrographs of biopsy specimens from a normal-appearing kidney (A) and from the kidney of patients with UPJ-O (B). Normal kidney does express high EGF mRNA, whereas in the renal tissue of patients with UJP-O there is a sharp decrease in the specific signal. EGF mRNA is localized mainly at the tubular level.

Fig 2. In situ hybridization for MCP-1. Bright-field photomicrographs of biopsy specimens from a normal-appearing kidney (A) and from the kidney of patients with UPJ-O (B). Normal kidney does not express MCP-1 mRNA, whereas in the renal tissue of patients with UPJ-O there is a sharp increase in the specific signal. MCP-1 mRNA is localized mainly at the interstitial level.

children with a history of UTI than in patients without UTI (MCP-1: UTI v non-UTI: 1.7 ⫾ 0.8 v 0.75 ⫾ 0.7 AU; P ⫽ .05). Moreover, EGF gene expression tended to be less in UTI patients compared with the non-UTI group (EGF: UTI v non-UTI: 1.2 ⫾ 1.5 v 2.1 ⫾ 1 AU; P value not significant). However, the analysis of variance performed on the 3 groups (no UTI, UTI, CTRL) for EGF gene expression resulted statistically significant (P ⬍ 0.01). Children with a history of UTI had a more severe prRT of the affected kidney compared with the non-UTI children (prRT: UTI v non-UTI: ⫺31% ⫾ 19 v ⫺7% ⫾ 19; P ⫽ .05). DF was reduced in UTI versus non-UTI cases, but differences were not statistically significant (DF: UTI v non-UTI 31% ⫾ 24 v 40% ⫾ 26; P value not significant).

located in the interstitium, but in most patients it can also be detected at glomerular and tubular levels. MCP-1 expression increases significantly in children with UTI, thus emphasizing the importance of UTI in the progression of renal damage. Furthermore, downregulation of EGF expression in UPJ-O kidneys was observed, which

DISCUSSION

Our results suggest the presence of a significant inflammatory response mediated by MCP-1/monocyte system in children with UPJ-O. This reaction mainly is

Fig 3. MCP-1 and EGF mRNA expression in control kidneys and in biopsy specimens from UPJ-O patients. MCP-1 and EGF mRNA signals were quantitated in all the biopsy specimens under investigation by 2 independent observers and expressed in arbitrary units on a 0 to 4 scale. Data are mean ⴞ SD. *P F .01 versus controls.

572

BARTOLI ET AL

Fig 4. Immunohistochemistry for CD68. Photomicrograph of biopsy specimens from UPJ-O patients. The arrows indicate CD68positive cells infiltrating the interstitial space.

may have important effects on subsequent renal growth and metabolism, thus confirming the experimental results obtained by other investigators.8 Our data show an inverse relationship between increased MCP-1 and reduced EGF mRNA expression suggesting a close connection between renal growth and inflammation. This hypothesis is supported by the significantly greater prRT of UPJ-O patients with UTI. So far we have not been able to show a direct relationship between age, differential function, and expression of EGF and MCP-1. The relationship between urinary tract obstruction and inflammation of renal parenchyma in the development of

glomerulosclerosis and tubular atrophy has been well documented in the literature.3 Our results suggest that an increased MCP-1 and a decreased EGF expression may play a key role in the basic mechanism that leads to the development of obstructive nephropathy. Monocytes seem to play a central role in the pathogenesis of the tubulointerstitial damage (tubular atrophy and necrosis) through the release of oxygen radicals and lysosomal enzymes13 regardless of the type of primary injury. Furthermore, monocytes may induce extracellular matrix deposition through the release of fibrogenic growth factors.14,15 In addition, several studies of animal13 and human9 models of primary or secondary tubulointerstitial damage have confirmed a direct relationship between these histopathologic changes and monocyte infiltration. The absence of a direct relationship between these pathological modifications and DF raises the issue of the sensitivity of this function in monitoring renal damage progression. Interestingly, rpRT may become an indirect index of renal growth failure because it seems to correlate with both decreased EGF and increased MCP-1 expression. Whether these pathological changes can be reversed by pyeloplasty only or by combined treatment (surgical plus medical) is currently being investigated in our laboratories. We believe that UPJ-O produces significant pathological changes into the affected kidney so that simple long-term observation may have deleterious and irreversible effects on renal parenchyma, especially in children with a history of UTI.

REFERENCES 1. Tapia J, Gonzalez R: Pyeloplasty improves renal function and somatic growth in children with ureteropelvic junction obstruction. J Urol 154:218-222, 1995 2. Koff SA, Campbell K: Nonoperative management of unilateral neonatal hydronephrosis. J Urol 148:525-531, 1992 3. Steinhardt GF, Ramon G, Salinas-Madrigal L: Glomerulosclerosis in obstructive uropathy. J Urol 140:1316-1318, 1988 4. Fisher DA, Salido EC, Barajas L: Epidermal growth factor and the kidney. Annu Rev Physiol 51:67-80, 1989 5. Harris RC: Potential physiologic roles for epidermal growth factor in the kidney. Am J Kidney Dis 17:627-630, 1991 6. Goodyer PR, Mulligan L, Goodyer CG: Expression of growthrelated genes in human fetal kidney. Am J Kidney Dis 17:608-610, 1991 7. Gesualdo L, Di Paolo S, Calabro` A, et al: Expression of epidermal growth factors and its receptor in normal and diseases human kidney: An immunohistochemical and in situ hybridization study. Kidney Int 49:656-665, 1996 8. Walton G, Buttyan R, Garcia-Montes E, et al: Renal growth factor expression during the early phase of experimental hydronephrosis. J Urol 148:510-514, 1992 9. Grandaliano G, Gesualdo L, Ranieri E, et al: Monocyte chemotac-

tic peptide-1 expression in acute and chronic nephritides: A pathogenetic role in interstitial monocytes recruitment. J Am Soc Nephrol 7:906-913, 1996 10. Diamond JR, Kees-Folts D, Ding G, et al: Macrophages, monocytes chemoattractant protein-1, and TGF-beta 1 in experimental hydronephrosis. Am J Physiol 266:F926-933, 1994 11. Rollins BJ, Stier P, Ernst et al: The human homolog of the JE gene encodes a monocyte secretory protein. Mol Cell Biol 509:46874695, 1989 12. Gesualdo L, Di Paolo S, Milani S, et al: Expression of platelet-derived growth factor receptors in normal and diseased human kidney.J Clin Invest 94:50-58, 1994 13. Eddy AA: Experimental insights into the tubulointerstitial disease accompanying primary glomerular lesions. J Am Soc Nephrol 5:1273-1287, 1994 14. Nathan CF: Secretory products of macrophages. J Clin Invest 79:319-326, 1987 15. Eddy AA, Li Z, McCulloch L: Macrophages may mediates interstitial fibrosis via the release of transforming growth factor beta in massive Heymann nephritis. J Am Soc Nephrol 2:573A, 1991