©

2 0 11 T H E A U T H O R S

1 3 3 6

B J U I N T E R N A T I O N A L

©

2 0 11 B J U I N T E R N A T I O N A L | 1 0 8 , 1 3 3 6 – 1 3 4 5 | doi:10.1111/j.1464-410X.2010.10074.x

2011 THE AUTHORS. BJU INTERNATIONAL 2011 BJU INTERNATIONALLaparoscopic and Robotic Urology

EMT IN HUMAN KIDNEYS CONTAINING LARGE CALCULIBOONLA

ET AL.

Fibrosis and evidence for epithelial-mesenchymal transition in the kidneys of patients with staghorn calculi

Chanchai Boonla, Kerstin Krieglstein*, Sombat Bovornpadungkitti

†

, Frank Strutz

‡

, Björn Spittau*, Chagkrapan Predanon

†

and Piyaratana Tosukhowong

Department of Biochemistry, Faculty of Medicine, Chulalongkorn University, Bangkok,

†

Unit of Urology, Surgical Department, Khon Kaen Hospital, Khon Kaen Province, Thailand, *Institute for Anatomy and Cell Biology, Department of Molecular Embryology, University of Freiburg, Freiburg, and

‡

Department of Nephrology and Rheumatology, Georg-August-University Medical Center, Göttingen, Germany

Accepted for publication 28 October 2010

RESULTS

• Overall, the kidney function of the patients was significantly reduced, indicated by increased plasma Cr and decreased corrected CCr compared with healthy controls.• Inflammation grading in renal tissues of the patients was correlated with the percentage of the fibrotic area. Renal fibrosis was inversely correlated with renal function.• Cytokeratins co-expressed with

α

SMA and vimentin were found in nephrolithiatic renal tubular cells, and these cells strongly expressed fibronectin and TGF-

β

1

.• Infiltration of CD68-positive cells was a common finding in the inflamed renal sections.

CONCLUSIONS

• Kidneys of large stone-forming patients had robust signs of inflammation and fibrosis, and there was a close correlation of renal fibrosis with renal dysfunction.• This is the first study to show evidence for renal tubular cells showing signs of EMT in large stone-containing kidneys. Plausibly, TGF-

β

1

triggers EMT, which at least in part contributes to large stone-induced renal fibrosis.

KEYWORDS

Kidney stone, EMT, renal fibrosis, inflammation, epithelial-mesenchymal transition, TGF-

β

1

What’s known on the subject? and What does the study add?

It has been shown that patients with nephrolithiasis did not have normal kidney function (retrospective study). Renal inflammation and fibrosis have been shown in kidney biopsies of nephrolithiasis patients, particularly those with brushite and cystine stones. No pathological change in kidney biopsies of patients with idiopathical calcium oxalate stone is noted.

Our cross-sectional study of patients with large kidney stone formation (mostly staghorn stones) shows that the patients have reduced overall kidney function regardless of stone type, and robust signs of inflammation and fibrosis are observed in stone-containing renal tissues. Reduced kidney function is closely associated with the degree of renal fibrosis. This is the first study demonstrating that renal tubular cells in stone-baring kidneys are undergoing epithelial-mesenchymal transition, and TGF-

β

1

is overproduced in these mesenchymalized cells.

Study Type – Therapy (case control)Level of Evidence 2b

OBJECTIVES

• To quantify fibrotic lesions in renal tissues obtained from patients with large calculi and to evaluate association with renal function.• Presence of epithelial-mesenchymal transition (EMT) in stone-containing renal tissues was investigated.

PATIENTS, SUBJECTS AND METHODS

• In all, 50 patients with nephrolithiasis with large calculi and matched healthy controls (37) were recruited.• Plasma creatinine (Cr) and corrected Cr clearance (CCr) were determined in all subjects.• Of the 50 patients, 38 had renal tissue available for histological analysis. Fibrosis was assessed by Masson’s trichrome staining. Co-expression of epithelial cytokeratins and mesenchymal markers [

α

-smooth muscle actin (

α

SMA) and vimentin] in renal tubular cells was detected by dual immunofluorescence staining.• Expression of fibronectin, transforming growth factor

β

1

(TGF-

β

1

) and CD68 were investigated.

BJUI

B J U I N T E R N A T I O N A L

E M T I N H U M A N K I D N E Y S C O N T A I N I N G L A R G E C A L C U L I

©

2 0 11 T H E A U T H O R S

B J U I N T E R N A T I O N A L

©

2 0 11 B J U I N T E R N A T I O N A L

1 3 3 7

INTRODUCTION

Kidney stone disease is endemic in Thailand, particularly in rural communities where hypocitraturia and hypokaliuria are prevalent [1–3]. The formation of stones in the kidneys and how the disease progresses are not fully understood, oxidative stress and inflammation are suggested as pathological pathways associated with kidney stone development [4–6]. Worcester

et al

. [7] reported decreased renal function in patients with nephrolithiasis. Evan

et al

. [8,9] reported inflammation and extensive renal fibrosis in various phenotypes of stone formers. We recently showed that patients with nephrolithiasis had intrarenal inflammation, and transcript expression of monocyte-chemoattractant protein1 (MCP-1) and interleukin6 (IL-6) was associated with renal impairment [4]. We propose that continuing oxidative renal injury and obstruction by crystals/stones cause intrarenal inflammation and fibrosis leading to nephron loss.

Renal fibrosis represents an uncontrolled wound-healing process, which inevitably occurs after chronic inflammation. The key cellular effector in renal fibrogenesis is the myofibroblast, which is generated from various sources, e.g. resident fibroblasts, circulating fibrocytes, endothelial cells via endothelial-mesenchymal transition, and renal tubular epithelial cells via epithelial-mesenchymal transition (EMT) [10–12]. The term EMT describes the sequential events of phenotypic changes from an epithelial to a mesenchymal phenotype. Growing evidence indicates that EMT plays a significant role in tumour progression and organ fibrosis [10]. The first evidence for EMT in renal tissues was shown in a study by Strutz

et al

. [13]. They found that renal tubular cells in murine fibrotic kidneys expressed fibroblast specific protein 1, suggesting phenotypic conversion of tubular epithelial cells. Yang and Liu [14,15] subsequently dissected key events during EMT in both cell culture and animal models and proposed four key steps for EMT including; loss of epithelial cell adhesion,

de novo

expression of

α

-smooth muscle actin (

α

SMA) and actin reorganization, disruption of tubular basement membrane, and increase in migration and invasion capability. Later, Iwano

et al

. [16] clearly showed that a substantial proportion of fibroblasts in fibrotic kidneys were derived from renal tubular cells. Although many EMT inducing factors have been characterized, the most

potent inducer that is capable of initiating and completing the four steps of EMT is TGF-

β

1

[15]. As

α

SMA is not expressed in normal renal tubular cells, co-expression of

α

SMA and epithelial cytokeratins in fibrotic kidneys is widely used as an indicator of EMT in the transitional stage [15,17–19]. Vimentin is widely used as mesenchymal marker, thus co-expression of vimentin and cytokeratins is used to identify mesenchymal cells derived from the tubular epithelium. The existence of EMT has been shown in inflammation-mediated renal diseases [17–21]. Although the existence of interstitial fibrosis has been shown in kidney biopsies of patients with nephrolithiasis [22], renal tubular EMT in these patients has not been investigated.

In the present study, we examined inflammation, fibrosis and tubular EMT in renal biopsies of patients with large kidney stones. Association of fibrosis with renal function and TGF-

β

1

renal expression were investigated.

PATIENTS AND METHODS

We recruited 50 patients with nephrolithiasis, admitted to Khon Kaen Hospital, Khon Kaen province, Thailand in 2008. All patients had positive plain abdominal film and/or IVU for calculi in the kidney. Patients with anomalous kidney and other urinary tract diseases i.e. horseshoe kidney, polycystic kidney, congenital VUR, neurogenic bladder and any malignancies were excluded. To compare kidney function 37 matched healthy individuals were recruited. Healthy condition was confirmed from direct interview and/or previous medical evaluation reports. Written-informed consent was obtained from all participants before specimen collection, and the research protocol was approved by the Ethical Committee, Faculty of Medicine, Chulalongkorn University, Bangkok.

Single 24-h urine samples were collected. In patients, urine was collected on the day before surgery. Renal function was estimated by plasma creatinine (Cr) and Cr clearance (CCr). The corrected CCr adjusted for a body surface area of 1.73 m

2

was calculated [4]. Stone specimens were collected during surgery for analysing stone composition by Fourier transform-infrared spectroscopy.

Renal biopsy specimens from the patients were collected (wedge-resection at surgery)

by an urologist (S.B.) in accordance with standard procedure. The biopsy was taken only for research purposes. All biopsied tissues were taken from stone-bearing kidneys, and both renal cortex and medulla near the stones (stone adjacent renal tissues) were collected. The biopsied tissue was immersed in 10% formalin buffer. Tissue processing was carried out according to routine histological protocol. Serial paraffin-embedded sections were cut at 4

μ

M

for histological analysis.

HISTOLOGICAL EXAMINATIONS

Light microscopy

Renal sections were stained with haematoxylin and eosin (H&E) for inflammation grading. Four levels of inflammatory extent were graded by a ‘blinded’ investigator as negative (no sign of inflammatory reaction), low (

+

), intermediate(

++

) and high (

+++

) degree of interstitial inflammation. Masson’s trichrome staining was used to determine fibrotic lesions in renal sections [20,23]. The percentage of fibrotic area relative to overall fibrosis in the section was evaluated under high-power magnification, and the amount of collagen deposit (stained in blue) was digitally calculated from at least five micrographs (F.S.).

Double immunofluorescence staining

Renal sections were deparaffinized, rehydrated, and antigens were unmasked in 0.1 M citrate buffer using microwaves. Sections were blocked for non-specific bindings in 10% normal goat serum (AbD Serotec) for 1 h, then incubated in a mixture of 1:500 mouse anti-cytokeratins (Clone MNF116, Dako) and 1:400 rabbit anti-

α

SMA (Zymed) antibodies or mouse anti-cytokeratins and 1:200 rabbit anti-vimentin (AbD Serotec) antibodies at 4

°

C, overnight. Incubation with 1:500 Alexa Fluor® 568 (red) goat anti-mouse and Alexa Fluor® 488 (green) goat anti-rabbit secondary antibodies (Molecular Probe) followed for 2 h in the dark. The stained sections were counterstained with 1:1000 4

′

-6-diamidino-2-phenylindole (blue for nuclei) for 5 min. Sections were mounted (Fluoromount-G, SouthernBiotech) and visualized under fluorescence microscope (Keyence, BZ-8000K). No cross-reactivity of anti-mouse Alexa-secondary antibody with rabbit primary antibody and anti-rabbit

B O O N L A

E T A L .

©

2 0 11 T H E A U T H O R S

1 3 3 8

B J U I N T E R N A T I O N A L

©

2 0 11 B J U I N T E R N A T I O N A L

Alexa-secondary antibody with mouse primary antibody was detected.

Immunoperoxidase staining

After antigen retrieval, renal sections were submerged in 0.1% H

2

O

2

in methanol for 30 min to exhaust endogenous peroxidase activity. Sections were incubated with 10% normal goat serum for 30 min, and then with 1:25 mouse anti-TGF-

β

1

(Clone 9016, R&D System) or 1:50 mouse anti-CD68 (Clone KP1, Dako) or 1:200 mouse anti-cellular fibronectin (Clone FN-3E2, Sigma) at 4

o

C, overnight, except for anti-TGF-

β

1

which was incubated for 48 h. After that, sections were incubated with 1:600 biotin-conjugated goat anti-mouse Ig (Vector Labs) for 2 h, followed by incubation with ABC reagent (Elite Vector Kit) for 2 h. Immunoreactive complexes were visualized with Vector SG substrate kit (SK4700, Vector Labs). For nuclei labelling, sections were counterstained with nuclear fast red.

STATISTICAL ANALYSIS

The data are presented as the mean (

SD

) or median (interquartile range, IQR) as appropriate. Two independent groups were compared by a two-sample

t

-test or Mann–Whitney

U

-test where appropriate. Differences among the three independent groups were assessed by a Kruskal–Wallis test. Correlation between two variables was determined by Spearman’s rank correlation test. A

P

<

0.05 was considered to indicate statistical significance.

RESULTS

In all, 50 patients with kidney stones, aged 51.66 (11.73) years (60% women) were recruited for the study (Table 1), and 37 age- [54.97 (11.61) years], gender-(78% women) and body mass index (BMI)-matched [23.43 (3.42) kg/m

2

] ‘healthy’ subjects were used for comparison of renal function. The mean age, sex distribution and BMI were not significantly different between the groups (Table 1). Of the 50 patients, stone specimens were obtained from 38 patients. Stones were categorized into four types according to the main crystal composition : calcium oxalate (CaOx, 21), calcium phosphate (CaP, five), uric acid (UA, nine) and magnesium ammonium phosphate (three). Urinary metabolic risk factors of patients with different stone types are shown in Table 2.

Renal biopsies were obtained from 38 patients, aged 51.28 (12.15) years (74%

women), and were used for histopathological study. Of these, stone types could be identified in 31 patients, i.e. 18 (58%) were CaOx, five (16%) CaP, seven (23%) UA, and one (3%) magnesium ammonium phosphate. One patient had had recurrent stones while it was the first stone episode for the rest. Only two patients underwent percutaneous nephrolithotomy whereas the remaining cases had open surgery for stone removal. Most of the patients (74%) had unilateral stones while the remaining cases had bilateral stones. Almost all patients (91%) developed staghorn stones. Formation of multiple stones occurred in 88% of the patients while single stones were found in 12%. Proportions of stones located in the calyx, pelvis and both sites were 3%, 12% and 85%, respectively. The mean stone size (estimated from plain abdominal film of kidney, ureter and bladder) was 6.0

×

3.6 cm diameter (largest 10.8

×

6.8 cm, smallest 2.4

×

2.0 cm), showing that the patients had large calculi.

TABLE 1

Renal function of patients with nephrolithiasis and healthy controls by gender

Clinical characteristicsHealthy NephrolithiasisTotal Men Women Total Men Women

N(%) 37 (100) 8 (22) 29 (78) 50 (100) 20 (40) 30 (60)Mean (

SD

):Age, years 54.97 (11.61) 60.88 (15.73) 53.34 (9.93) 51.66 (11.73) 55.89 (12.57) 48.79 (10.38)BMI,kg/m

2

23.43 (3.42) 26.56 (4.86) 22.57 (2.36) 24.09 (5.09) 23.14 (3.62) 24.61 (5.75)Median (IQR)

Urine volume, mL 1640 (1265) 1585 (2470) 1640 (1045) 1200 (900) 1200 (825) 1200 (900)Plasma Cr, mg/dL 0.86 (0.24) 1.11 (0.10) 0.84 (0.07) 1.14 (0.40)* 1.15 (0.34) 1.04 (0.42)‡CCr, mL/min 75.71 (47.07) 120.00 (95.45) 68.88 (37.17) 36.07 (49.51)* 24.39 (43.60)† 39.54 (49.91)‡Corrected CCr,mL/min/1.73 m

2

83.91 (45.84) 91.01 (97.94) 81.21 (39.59) 35.12 (52.18)* 32.19 (52.18)† 35.12 (58.61)‡

*

P

<

0.05 vs total healthy; †

P

<

0.05 vs healthy men; ‡

P

<

0.05; vs healthy females.

TABLE 2

Urinary metabolic profiles of patients with different stone types

Urinary variablesStone typesCaOx CaP UA MAP

Median (IQR):Calcium, mg/day 42.20 (79.33) 49.98 (27.84) 70.88 (90.79) 18.86 (17.29)Oxalate, mg/day 10.68 (76.05) 10.43 (5.85) 8.30 (13.08) NDPhosphate, mg/day 124.14 (112) 234.86 (125.15) 152.57 (225.50) 125.14 (111.59)Uric acid, mg/day 409.06 (435.24) 391.03 (98.12) 339.66 (280.34) 480.00 (561.86)Potassium, mEq/day 17.20 (20.90) 13.20 (2.40) 13.45 (7.90) 12.00 (29.45)Citrate, mg/day 101.90 (76.05) 73.10 (78.78) 144.45 (101.10) 95.19 (143 012)

MAP, magnesium ammonium phosphate; ND, not determined;

P

>

0.05 (one-way

ANOVA

) for all urinary variables.

E M T I N H U M A N K I D N E Y S C O N T A I N I N G L A R G E C A L C U L I

©

2 0 11 T H E A U T H O R S

B J U I N T E R N A T I O N A L

©

2 0 11 B J U I N T E R N A T I O N A L

1 3 3 9

Urinary tract obstruction and infections were classified according to IVU results and pyuria (urinary white blood cells

>

5 cells/high-power field), respectively. Of 38 patients, 30 had IVU data and 31 were tested for pyuria. Four (13%) patients had bilateral hydronephrosis (defined as urinary obstruction), 17 (57%) had unilateral hydronephrosis, and nine (30%) had no evidence of hydronephrosis. In all, 22 patients (71%) had pyuria while ninecases (29%) had no pyuria.

The above clinical data show that our studied cohort had a severe form of kidney stone disease, which is common in rural communities of Thailand. In all, 58 renal biopsies from 38 patients were available for histological evaluation and 20 patients had two renal biopsies for investigation. Control renal sections were obtained from non-pathological regions of kidneys removed due to tumours.

IMPAIRMENT OF RENAL FUNCTION

The patients with calculi had significantly higher plasma Cr levels than the healthy controls (Table 1, Fig. 1a). Corrected CCr of the patients was also significantly lower than that of controls (Table 1, Fig. 1b). As we accounted for the influence of sex on plasma Cr and CCr by performing sub-analysis, female patients had significantly greater plasma Cr than female controls (Table 1). Both CCr and corrected CCr of female patients were significantly lower than those of healthy females. There was no significant increase in plasma Cr level in the male patients. However, there were significant reductions in CCr and corrected CCr in male patients (Table 1). Our data clearly show that kidney function of large stone-forming patients is impaired. We further asked if the impairment of their kidneys was associated with intrarenal inflammation and fibrosis.

Among patients with different stone types, levels of plasma Cr were not significantly different (Fig. 1c). Likewise, there were no significant differences in corrected CCr among the different types of stones (Fig. 1d).

However, patients with urinary obstruction (bilateral hydronephrosis) had significantly higher plasma Cr (

P

=

0.026) and a trend of decreased corrected CCr (

P

=

0.055), compared with those without bilateral obstruction.

RENAL INFLAMMATION, FIBROSIS AND RENAL DYSFUNCTION

There were no signs of inflammation and fibrosis in control renal sections, as shown by H&E staining (Fig. 2a). By contrast, infiltration of mononuclear cells and sclerotic appearance were often found in renal sections from patients with calculi (Fig. 2b). Degrees of inflammation ranged from mild to severe. Glomerulosclerosis was common in renal sections with high degrees of inflammation (Fig. 2b). The basal level of collagen deposits (a hallmark of renal fibrosis) in control renal tissues is shown in Fig. 2c and the level was markedly increased in renal tissues of patients with calculi (Fig. 2d). There was increased collagen depositsin both tubulointerstitium and fibrotic glomeruli indicating tubulointerstitial fibrosis and glomerulosclerosis, respectively. We therefore concluded that there were inflammation and fibrosis in the kidneys containing large calculi.

All renal sections from the patients with calculi had some degree of inflammation. There was a significant positive correlation between inflammation grading and fibrotic area (Spearman’s

ρ

0.301,

P

=

0.023). Sections with high inflammation also had a high percentage of fibrotic area (Fig. 3a). Although there was no statistical significance, there were trends of associations of increased intrarenal inflammation with elevated plasma Cr and reduced corrected CCr (Fig. 3b). Importantly, fibrotic area was linearly correlated with plasma Cr level with a Spearman’s

ρ

of 0.416 (

P

=

0.014; Fig. 3c). Furthermore, the relative percentage of the fibrotic area was inversely correlated with corrected CCr (Spearman’s

ρ

–0.539,

P

=

0.002; Fig. 3d). Thus, our data indicate a close link between renal fibrosis and renal dysfunction in patients with large kidney calculi.

We assessed the relationship between pyuria and intrarenal inflammation. The intrarenal inflammation was re-categorized into low (no inflammation and

+

) and high (

++

and

+++

)

FIG. 1.

Plasma Cr levels and corrected CCr compared between healthy subjects and patients with nephrolithiasis (

a

,

b

), as well as among patients with different stone types (

c

,

d

). Plasma Cr levels in patients with nephrolithiasis was significantly higher than that in healthy controls (

a

). There was a significant reduction of corrected CCr in patients with nephrolithiasis (

b

). When comparing patients with different stone types, neither plasma Cr (

c

) nor corrected CCr (

d

) were significantly different. CaOx (21), CaP (five), UA(nine), magnesium ammonium phosphate (MAP, three).

Plas

ma

Cr, m

g/dl

Corr

ecte

d CC

r, m

l/min

/1.7

3 m

2

Plas

ma

Cr, m

g/dl

Corr

ecte

d CC

r, m

l/min

/1.7

3 m

2

Healthy Nephro-lithiasis

Healthy Nephro-lithiasis

CaOx CaP UA MAP CaOx CaP UA MAP

P < 0.001 P < 0.001

P = 0.723P = 0.967

2.0

1.5

1.0

0.5

2.0

1.5

1.0

0.5

150

100

50

0

150

100

50

0

a b

c d

FIG. 2.

H&E (

a

,

b

) and Masson’s trichrome (

c

,

d

) staining of control and nephrolithiasis renal tissues. H&E micrograph of renal control section showed no or few infiltrating cells (

a

). There was a marked infiltration of mononuclear cells in renal sections from patients with nephrolithiasis (

b

). Glomerulosclerosis (arrows) was commonly seen in sections with marked leukocyte infiltration. Masson’s trichrome-stained control renal sections showed a basal level of collagen deposit (blue;

c

). Extensive positive Masson’s trichrome staining in renal interstitium of nephrolithiasis renal sections was commonly found (

d

). Inset shows fibrotic glomerulus in nephrolithiasis renal sections positive for Masson’s trichrome staining.

a

and

b

×

100;

c

and

d

×

400.

B O O N L A

E T A L .

©

2 0 11 T H E A U T H O R S

1 3 4 0

B J U I N T E R N A T I O N A L

©

2 0 11 B J U I N T E R N A T I O N A L

inflammation. Using Fisher’s exact test, patients with pyuria (22, 71%) tended to have a high degree of inflammation, although it was not statistically significant (

P

=

0.056). In all, 16 of 22 patients (73%) with pyuria presented with high intrarenal inflammation. Conversely, six of nine patients without pyuria had low intrarenal inflammation. There was no significant association between urinary obstruction (bilateral hydronephrosis) and degree of intrarenal inflammation (

P

=

1.000). However, all four patients with bilateral obstruction had a high degree of intrarenal inflammation. Associations of renal fibrosis with pyuria and urinary obstruction were also

assessed. Neither pyuria (

P

=

0.796) nor obstruction status (

P

=

0.077) was significantly associated with percentage of fibrotic area in renal sections of the patients, as evaluated by Mann–Whitney test.

Of the 38 patients included in the histopathological investigation, 31 had stones available for stone-type analysis. CaOx (18), CaP (five), UA (seven) and magnesium ammonium phosphate (one) stones accounting for 58%, 16%, 23% and 3%, respectively. The renal section of the patient with the magnesium ammonium phosphate stone had high inflammation (

+++

), and its

fibrotic area was greater than the average fibrotic areas of the other stone types. Although the data was drawn from only one magnesium ammonium phosphate case, it might imply that patients with infection stones have agreater extent of intrarenal inflammation and fibrosis compared with non-infection stones. We further compared extents of intrarenal inflammation and fibrosis among patients with non-infection stones (CaOx, CaP and UA). Based on Fisher’s exact test, there was no significant association between these three stone types and the degree of intrarenal inflammation (

P

=

0.871). Similarly, the percentage of

FIG. 3.

Associations of renal fibrosis with inflammation and renal dysfunction in patients with nephrolithiasis. Renal sections with larger fibrotic areas had a higher degree of inflammation (semi-quantitative scoring; Kruskal–Wallis test,

P

=

0.021) (

a

). Increased intrarenal inflammation tended to be associated with increased plasma Cr and decreased corrected CCr (

b

). Increased fibrotic area was significantly correlated to increased plasma Cr (

c

). There was a significant inverse correlation between renal fibrosis and corrected CCr (

d

).

a b

dc

Fibr

otic

are

a, %

Plas

ma

Cr, m

g/dl

Corr

ecte

d CC

r, m

l/min

/1.7

3 m

2

Fibr

otic

are

a, %

Fibr

otic

are

a, %

Low Intermediate High

Low Interm. High Low Interm. High

Inflammation

Inflammation

Plasma Cr, mg/dl Corrected CCr, ml/min/1.73 m2

50

40

30

20

10

1.6

1.4

1.2

1.0

0.8

0.6

200

150

100

50

0

50.00

40.00

30.00

20.00

10.00

0.00

50.00

40.00

30.00

20.00

10.00

0.000.50 1.00 1.50 2.00 2.50 3.00 3.50 0.00 50.00 100.00 150.00 200.00 250.00

Spearman’s rho = 0.416, P = 0.014 Spearman’s rho = -0.539, P = 0.002

E M T I N H U M A N K I D N E Y S C O N T A I N I N G L A R G E C A L C U L I

©

2 0 11 T H E A U T H O R S

B J U I N T E R N A T I O N A L

©

2 0 11 B J U I N T E R N A T I O N A L

1 3 4 1

fibrotic area among these three groups of patients did not vary (Kruskal–Wallis test, P = 0.883). The present findings show that the extent of intrarenal inflammation and fibrosis among patients with CaOx, CaP and UA stones were not significantly different. It has been

stated that renal fibrosis is common in inflammatory-mediated renal diseases, and from a certain point the fibrogenic pathway becomes independent of the disease aetiology [18]. This could explain the comparable extent of renal fibrosis among CaOx, CaP and

UA stones, while they had different aetiologies.

SIGNS OF RENAL TUBULAR EMT

In control renal tissues, there was no expression of αSMA (Fig. 4b) and vimentin (Fig. 4e) in renal tubular cells. The cytokeratins antibody effectively labelled renal tubular epithelial cells in both control (Fig. 4a,d) and patient renal sections (Fig. 4g,j). Some renal tubules in the sections from patients were faintly positive for cytokeratins, but strongly positive for αSMA and vimentin. This may indicate a complete loss of epithelial phenotype and gain of mesenchymal characteristics. αSMA and vimentin were strongly expressed in renal tubular cells of the patients (Fig. 4h,k). Co-expression of cytokeratins and αSMA as well as vimentin was also seen in medullary renal tubular cells (Fig. 4m). We detected interstitial cells that were positive for cytokeratins, αSMA (Fig. 4n) and vimentin (Fig. 4o), indicating a late stage of renal tubular EMT. We further investigated if cells undergoing EMT were able to produce matrix proteins. There was increased expression of cellular fibronectin in these mesenchymalized tubular cells (Fig. 5b), whereas there was only very low or even negative expression of cellular fibronectin in renal tubular cells of control sections (Fig. 5a). Based on our present findings, renal tubular EMT exists in the renal tissues of patients with large calculi and these tubular cells are capable of producing matrix proteins.

TGF-β1 AND CD68 EXPRESSIONS

TGF-β1 staining was negative (or very slightly positive in some tubules) in renal tubular cells of control sections (Fig. 5c). There was increased TGF-β1 expression in renal sections of patients with calculi (Fig. 5d). TGF-β1 was strongly expressed in renal tubular cells (Fig. 5e). Using Fisher’s exact test there was a significant correlation of positive TGF-β1 expression and the degree of inflammation (P = 0.043). However, there was no significant correlation between TGF-β1 expression and the percentage of fibrotic area. There was negative renal expression of TGF-β1 in 29% (11/38) of the patients. Interstitial mononuclear cells were also positive for TGF-β1 (Fig. 4e), and these cells might be infiltrated monocytes and macrophages or mesenchymalized tubular cells. Based on our data, renal tubular cells in nephrolithiatic kidneys over-produced TGF-β1.

FIG. 4. Double immunofluorescence staining to show co-expression of an epithelial marker (cytokeratins; red) and mescenchymal markers (αSMA and vimentin; green) in renal biopsies of patients with nephrolithiasis. In control renal tissues (a–f), there were no or only slight expression of αSMA and vimentin. There was strong expression of cytokeratins in unaltered renal tubular cells in control sections (a, d). In micrographs g–l (serial sections), as representatives of nephrolithiasis renal tissues, co-localizations of cytokeratins and αSMA, and cytokeratins and vimentin (yellow) in renal tubular cells were frequently found. Micrograph m shows co-expression of cytokeratins and αSMA in medullary renal tubules. The co-expression of epithelial and mesenchymal markers indicates the transitional stage of EMT. Micrographs n and o, serial renal sections from patients with nephrolithiasis, show an interstitial cell expressing cytokeratins, αSMA and vimentin (arrows), defined as an EMT transdifferentiating cell. Nuclei were stained with 4′-6-diamidino-2-phenylindole (blue). ×400.

B O O N L A E T A L .

© 2 0 11 T H E A U T H O R S

1 3 4 2 B J U I N T E R N A T I O N A L © 2 0 11 B J U I N T E R N A T I O N A L

Very rare CD68-positive cells (marker for monocytes/macrophages) were found in the renal interstitium of control sections (Fig. 5f). By contrast, CD68-positive cells were commonly found in the renal sections of patients with calculi (Fig. 5g,h). Additionally, expression of CD68 in renal tissues of the patients correlated well with the degree of inflammation.

DISCUSSION

Decreased renal function has been reported in patients with nephrolithiasi seven in the mildest form (idiopathic CaOx stone) [7]. In the present study, there was an impairment of kidney function in patients with large calculi, as indicated by increased plasma Cr and decreased corrected CCr. It is well accepted that large stone formation (mostly staghorn stones) in the kidneys can obstruct urine passage, cause chronic inflammation, and eventually destroy the kidneys [24]. Large or staghorn stones are rarely found in the Western world, but are still prevalent in developing countries, for instance in rural communities of Thailand. Impairment of overall kidney function in the present patients with large calculi might be the result of chronic inflammation induced by long-term stone formation, obstruction as well as infection. However, most of the patients had unilateral stones. Impaired kidney function in patients with unilateral stones is rarely reported. Renal dysfunction can be secondary to potassium depletion [25,26]. Our previous data showed that patients with renal calculi in rural communities of north-eastern Thailand (patients in the present study were also from the Northeast) were in a state of potassium deficiency [27,28], caused by low potassium intake and high sweat loss [29]. In the present study, we did not measure serum potassium, but hypokaliuria was common in our patients [Table 2; median (IQR) urinary potassium 13.2 (11.0) mEq/day, reference >30 mEq/day). Additionally, hypocitraturia is highly prevalent in Thai patients with nephrolithiasis (Table 2) [2,3], which has been suggested to be a consequence of potassium depletion [30]. Therefore, impaired kidney function in the present patients, particularly those with unilateral stones, might be partly caused by potassium depletion.

Increasing evidence shows that kidney stones increase the risk of chronic kidney disease and end-stage renal disease [31–35]. However,

FIG. 5. Representative immunohistostainings for fibronectin, TGF-β1 and CD68 (ED1 antigen). Fibronectin was rarely expressed in renal tubular cells of control renal sections (a). In nephrolithiasis sections, fibronectin was strongly expressed in renal tubular cells, interstitial fibroblast-like cells (white arrow), and interstitial mononuclear cells (black arrow) (b). Expression of TGF-β1 in control renal sections was low or even negative (c). By contrast, TGF-β1 was over-expressed in nephrolithiasis renal tissues (d, e). With higher magnification (e), TGF-β1 was mainly positive in renal tubular cells. However, there were also TGF-β1-expressing cells in the renal interstitium (arrow). CD68-positive cells (monocytes/macrophages) were rarely found in the renal interstitium of control sections (f), but they were commonly found in blood vessels serving as internal positive controls (arrows); inset, ×400. Increased infiltration of CD68-positive cells was frequently found in renal sections of patients with nephrolithiasis, particularly in sections with widespread inflammation (g). Most of CD68-positive cells resided in the renal interstitial space (h), defined as infiltrating monocytes/macrophages. Specific reactivity (blue-grey). Nuclei (red). d, f, g ×100 and a, b, c, e, h ×400.

E M T I N H U M A N K I D N E Y S C O N T A I N I N G L A R G E C A L C U L I

© 2 0 11 T H E A U T H O R S

B J U I N T E R N A T I O N A L © 2 0 11 B J U I N T E R N A T I O N A L 1 3 4 3

direct evidence of stone-induced renal inflammation and fibrosis leading to progressive nephron loss has not been provided. Renal histopathology of various types of nephrolithiasis has been intensively investigated by Evan et al. [22,36–38]. Randall’s plaques are abundantly found in renal papillary of idiopathic CaOx stone formers (n = 15), but there are no signs of inflammation and fibrosis [39]. In patients with brushite stones (n = 10) [9], obesity bypass (CaOx, n = 4) [39], cystine stones (n = 7) [8], distal renal tubular acidosis (apatite, n = 5) [40], primary hyperparathyroid (CaOx and apatite, n = 5) [41], and ileostomy (CaOx and UA, n = 7) [42], medullary collecting ducts and ducts of Bellini are usually plugged with the relevant crystals, and papillary flattening and atrophy, epithelial loss, inflammation, interstitial fibrosis and glomerulosclerosis, particularly around the plugged ducts, are documented. We recently reported increased intrarenal expressions of MCP-1 and IL-6 mRNAs in nephrolithiasis renal tissues, and their levels were associated with renal dysfunction [4]. In the present study, there were variable degree of inflammation and fibrosis in renal sections of patients with large stone formations. As almost all of the studied patients had staghorn stones, obstruction was expected (70% had unilateral or bilateral hydronephrosis). The percentage fibrotic area was well correlated with the extent of inflammation. Importantly, increased fibrotic extent in stone-baring kidneys was significantly associated with a decline of whole kidney function. Our human data strongly support that renal fibrosis closely associates with the progressive loss of nephrons in patients with large kidney calculi. Effective approach to recuperate kidney function is recommended for the postoperative management of these patients.

Some discrepancies with the findings of Evan et al. [39] should be mentioned. In the present study, there was inflammation and sclerotic appearance in all renal tissues of patients CaOx stones (n = 18). The main reason may be stone burden, as all of the present patients had large staghorn stone formations with mean size of 6.0 × 3.6 cm. Of the 18 patients with CaOx stones, one had a pure CaOx stone while the stones in the remaining cases were mixed with CaP or magnesium ammonium phosphate. Pyuria and obstruction were commonly detected in the present cohort. Due to a large number of patients with nephrolithiasis and the limited number of

fully-equipped hospitals in Thailand, the patients usually have to wait for their operation for at least 3–6 months after diagnosis, thus causing prolonged urinary obstruction and infection. Therefore, the severe renal pathology in the present patients was possibly caused by obstruction, infection and chronic inflammation, resulting from large stone mass and long stone-forming periods. We performed Yasue’s staining to elucidate the deposit of calcium-containing crystals in all renal sections, but only one section was positive (data not shown). As formalin-fixed, paraffin-embedded renal sections were used for Yasue’s staining, we thought that the calcium-containing crystals might have been lost during the routine tissue preparation as suggested in the study by Kummeling et al. [43].

Conversion of renal tubular epithelial cells via EMT is a significant source of myofibroblasts in fibrotic kidneys [11,16,44]. In the present study, we investigated whether renal tubular cells in large stone-containing fibrotic kidneys were undergoing EMT. In all nephrolithiasis renal sections, there was co-expression of epithelial type cytokeratins with mesechymal markers (αSMA and vimentin) in renal tubular cells. Moreover, fibronectin was over-expressed in these cells. The findings indicated that renal tubular cells in large stone-containing kidneys were undergoing EMT, and these cells were capable of synthesizing matrix proteins. We found interstitial cells that expressed cytokeratins, αSMA and vimentin in many patients’ renal sections (Fig. 4n,o), suggested as a sign of tubular cells migration toward the interstitium. We considered these cells as mesenchymalized tubular cells in the last stage of EMT [19]. Tubular EMT has been shown in renal tissues of patients with membranous nephropathy [18], IgA nephropathy [20], lupus nephritis [21] and chronic allograft nephropathy [17,19]. To our knowledge, this is the first report of evidence for tubular EMT in kidney tissues of patients with large stone formations. We suggest that this phenotypic conversion pathway actively contributes to large stone-induced renal fibrosis.

TGF-β1 is a well-characterized profibrotic factor that plays a central role in renal fibrogenesis [11]. Over-expression of TGF-β1 in fibrotic renal diseases is well recognized [45]. Increased production of TGF-β1 in oxalate-treated renal tubular (LLC-PK1) cells has been shown [46]. However, there has been no

report of TGF-β1 expression in renal tissues of patients with renal calculi. The present data show that mesenchymalized tubular cells in stone-baring fibrotic kidneys over-expressed TGF-β1. We hypothesize that crystals/stones and obstruction induce TGF-β1 production in renal tubular cells, and the produced TGF-β1 in turn triggers EMT both in autocrine and paracrine manners. However, this hypothesis needs further experimental proof. It should be noted that TGF-β1 was not over-expressed in all nephrolithiatic renal tissue. There was negative renal expression of TGF-β1 in 11 patients (29%), although there were fibrotic lesions in their kidneys. There are three isoforms of TGF-β (-β1, -β2 and -β3), and they are shown to have relatively similar fibrogenic effect. In addition, TGF-β1 over-expression does not dominate in all fibrogenic systems [47]. As the antibody used in the present study recognized only TGF-β1, it might be possible that other isoforms predominated in the TGF-β1-negative renal tissues.

Crystal-induced renal inflammation in nephrolithiasis has been proposed [6], and marked infiltration of ED1-positive inflammatory cells in renal interstitium of nephrolithic rats was demonstrated [48]. As expected, we found infiltrating CD68-positive mononuclear cells in inflamed renal sections obtained from patients with kidney calculi, supporting the role of inflammatory cells in kidney stone development.

A limitation of the present study should be mentioned. It is known that chronic potassium depletion can alter function and structure of the kidney [25,26]. The serum potassium of the present patients was not available. As potassium depletion is commonly observed in Thai renal stone formers, renal tissues used in the present study might represent renal tissues of patients with hypokalaemia nephrolithiasis.

In conclusion, there was inflammation and fibrosis in the kidneys of the present patients with large staghorn stones. There was increased monocytes/macrophages recruitment in their renal tissues, and a close association of enhanced renal fibrosis in stone-baring kidneys with worse overall renal function. The renal tubular cells in large stone-containing kidneys underwent EMT, and TGF-β1 was over-expressed in these mesenchymalized tubular cells. We hypothesize that TGF-β1-induced tubular EMT contributes to large stone-induced renal

B O O N L A E T A L .

© 2 0 11 T H E A U T H O R S

1 3 4 4 B J U I N T E R N A T I O N A L © 2 0 11 B J U I N T E R N A T I O N A L

fibrosis, and the blockade of this alteration could be a novel therapeutic approach for patients with large calculi.

ACKNOWLEDGEMENTS

C.B. awarded a Humboldt research fellowship (Alexander von Humboldt Foundation). Ratchadapiseksompotch Research Fund (RA26/51(1) to P.T.). Partially supported by Thailand Research Fund (MRG5180159) and Asahi Foundation Oversea Grant (to C.B.). We thank Uraiwan Waiwijit and Sarawut Saepoo for their excellent assistant. In the memory of Dr. Sombat Bovornpadungkitti, an urologist who devoted almost all of his time for treating stone patients in rural communities of northeastern Thailand.

CONFLICT OF INTEREST

None declared.

REFERENCES

1 Tosukhowong P, Boonla C, Ratchanon S et al. Crystalline composition and etiologic factors of kidney stone in Thailand : update 2007. Asian Biomed 2007; 1: 87–95

2 Tosukhowong P, Yachantha C, Sasivongsbhakdi T et al. Citraturic, alkalinizing and antioxidative effects of limeade-based regimen in nephrolithiasis patients. Urol Res 2008; 36: 149–55

3 Tungsanga K, Sriboonlue P, Futrakul P, Yachantha C, Tosukhowong P. Renal tubular cell damage and oxidative stress in renal stone patients and the effect of potassium citrate treatment. Urol Res 2005; 33: 65–9

4 Boonla C, Hunapathed C, Bovornpadungkitti S et al. Messenger RNA expression of monocyte chemoattractant protein-1 and interleukin-6 in stone-containing kidneys. BJU Int 2008; 101: 1170–7

5 Boonla C, Wunsuwan R, Tungsanga K, Tosukhowong P. Urinary 8-hydroxydeoxyguanosine is elevated in patients with nephrolithiasis. Urol Res 2007; 35: 185–91

6 Khan SR. Crystal-induced inflammation of the kidneys: results from human studies, animal models, and tissue-culture studies. Clin Exp Nephrol 2004; 8: 75–88

7 Worcester EM, Parks JH, Evan AP, Coe FL. Renal function in patients with nephrolithiasis. J Urol 2006; 176: 600–3

8 Evan AP, Coe FL, Lingeman JE et al. Renal crystal deposits and histopathology in patients with cystine stones. Kidney Int 2006; 69: 2227–35

9 Evan AP, Lingeman JE, Coe FL et al. Crystal-associated nephropathy in patients with brushite nephrolithiasis. Kidney Int 2005; 67: 576–91

10 Zeisberg M, Kalluri R. The role of epithelial-to-mesenchymal transition in renal fibrosis. J Mol Med 2004; 82: 175–81

11 Liu Y. Renal fibrosis: new insights into the pathogenesis and therapeutics. Kidney Int 2006; 69: 213–7

12 Zeisberg EM, Potenta SE, Sugimoto H, Zeisberg M, Kalluri R. Fibroblasts in kidney fibrosis emerge via endothelial-to-mesenchymal transition. J Am Soc Nephrol 2008; 19: 2282–7

13 Strutz F, Okada H, Lo CW et al. Identification and characterization of a fibroblast marker: FSP1. J Cell Biol 1995; 130: 393–405

14 Yang J, Liu Y. Dissection of key events in tubular epithelial to myofibroblast transition and its implications in renal interstitial fibrosis. Am J Pathol 2001; 159: 1465–75

15 Liu Y. New insights into epithelial-mesenchymal transition in kidney fibrosis. J Am Soc Nephrol 2010; 21: 212–22

16 Iwano M, Plieth D, Danoff TM, Xue C, Okada H, Neilson EG. Evidence that fibroblasts derive from epithelium during tissue fibrosis. J Clin Invest 2002; 110: 341–50

17 Hertig A, Verine J, Mougenot B et al. Risk factors for early epithelial to mesenchymal transition in renal grafts. Am J Transplant 2006; 6: 2937–46

18 Rastaldi MP, Ferrario F, Giardino L et al. Epithelial-mesenchymal transition of tubular epithelial cells in human renal biopsies. Kidney Int 2002; 62: 137–46

19 Vongwiwatana A, Tasanarong A, Rayner DC, Melk A, Halloran PF. Epithelial to mesenchymal transition during late deterioration of human kidney transplants: the role of tubular cells in fibrogenesis. Am J Transplant 2005; 5: 1367–74

20 Nishitani Y, Iwano M, Yamaguchi Y et al. Fibroblast-specific protein 1 is a

specific prognostic marker for renal survival in patients with IgAN. Kidney Int 2005; 68: 1078–85

21 Rossini M, Cheunsuchon B, Donnert E et al. Immunolocalization of fibroblast growth factor-1 (FGF-1), its receptor (FGFR-1), and fibroblast-specific protein-1 (FSP-1) in inflammatory renal disease. Kidney Int 2005; 68: 2621–8

22 Evan AP, Coe FL, Lingeman JE, Worcester E. Insights on the pathology of kidney stone formation. Urol Res 2005; 33: 383–9

23 Moriyama T, Kawada N, Ando A et al. Up-regulation of HSP47 in the mouse kidneys with unilateral ureteral obstruction. Kidney Int 1998; 54: 110–9

24 Healy KA, Ogan K. Pathophysiology and management of infectious staghorn calculi. Urol Clin North Am 2007; 34: 363–74

25 Okun R, Kleeman CR. Renal disease secondary to metabolic disorders or physiological deficiency states. Calif Med 1967; 107: 8–10

26 Weiner ID, Wingo CS. Hypokalemia – consequences, causes, and correction. J Am Soc Nephrol 1997; 8: 1179–88

27 Sriboonlue P, Prasongwattana V, Tungsanga K et al. Blood and urinary aggregator and inhibitor composition in controls and renal-stone patients from northeastern Thailand. Nephron 1991; 59: 591–6

28 Tungsanga K, Sriboonlue P, Borwornpadungkitti S, Tosukhowong P, Sitprija V. Urinary acidification in renal stone patients from northeastern Thailand. J Urol 1992; 147: 325–8

29 Sriboonlue P, Prasongwatana V, Suwantrai S, Bovornpadungkitti S, Tungsanga K, Tosukhowong P. Nutritional potassium status of healthy adult males residing in the rural northeast Thailand. J Med Assoc Thai 1998; 81: 223–32

30 Yachantha C, Hossain RZ, Yamakawa K et al. Effect of potassium depletion on urinary stone risk factors in Wistar rats. Urol Res 2009; 37: 311–6

31 Jungers P, Joly D, Barbey F, Choukroun G, Daudon M. ESRD caused by nephrolithiasis: prevalence, mechanisms, and prevention. Am J Kidney Dis 2004; 44: 799–805

32 Rule AD, Bergstralh EJ, Melton LJ 3rd, Li X, Weaver AL, Lieske JC. Kidney stones and the risk for chronic kidney

E M T I N H U M A N K I D N E Y S C O N T A I N I N G L A R G E C A L C U L I

© 2 0 11 T H E A U T H O R S

B J U I N T E R N A T I O N A L © 2 0 11 B J U I N T E R N A T I O N A L 1 3 4 5

disease. Clin J Am Soc Nephrol 2009; 4: 804–11

33 Stankus N, Hammes M, Gillen D, Worcester E. African American ESRD patients have a high pre-dialysis prevalence of kidney stones compared to NHANES III. Urol Res 2007; 35: 83–7

34 Vupputuri S, Soucie JM, McClellan W, Sandler DP. History of kidney stones as a possible risk factor for chronic kidney disease. Ann Epidemiol 2004; 14: 222–8

35 Saucier NA, Sinha MK, Liang KV et al. Risk factors for CKD in persons with kidney stones: a case-control study in Olmsted County, Minnesota. Am J Kidney Dis 2010; 55: 61–8

36 Evan AP. Physiopathology and etiology of stone formation in the kidney and the urinary tract. Pediatr Nephrol 2010; 25: 831–41

37 Evan AP, Lingeman JE, Coe FL, Worcester EM. Role of interstitial apatite plaque in the pathogenesis of the common calcium oxalate stone. Semin Nephrol 2008; 28: 111–9

38 Evan A, Lingeman J, Coe FL, Worcester E. Randall’s plaque: pathogenesis and role in calcium oxalate nephrolithiasis. Kidney Int 2006; 69: 1313–8

39 Evan AP, Lingeman JE, Coe FL et al. Randall’s plaque of patients with

nephrolithiasis begins in basement membranes of thin loops of Henle. J Clin Invest 2003; 111: 607–16

40 Evan AP, Lingeman J, Coe F et al. Renal histopathology of stone-forming patients with distal renal tubular acidosis. Kidney Int 2007; 71: 795–801

41 Evan AE, Lingeman JE, Coe FL et al. Histopathology and surgical anatomy of patients with primary hyperparathyroidism and calcium phosphate stones. Kidney Int 2008; 74: 223–9

42 Evan AP, Lingeman JE, Coe FL et al. Intra-tubular deposits, urine and stone composition are divergent in patients with ileostomy. Kidney Int 2009; 76: 1081–8

43 Kummeling MT, de Jong BW, Laffeber C et al. Tubular and interstitial nephrocalcinosis. J Urol 2007; 178: 1097–103

44 Strutz F. How many different roads may a cell walk down in order to become a fibroblast? J Am Soc Nephrol 2008; 19: 2246–8

45 Huang Y, Border WA, Noble NA. Perspectives on blockade of TGFbeta overexpression. Kidney Int 2006; 69: 1713–4

46 Rashed T, Menon M, Thamilselvan S. Molecular mechanism of oxalate-induced

free radical production and glutathione redox imbalance in renal epithelial cells: effect of antioxidants. Am J Nephrol 2004; 24: 557–68

47 Yu L, Border WA, Huang Y, Noble NA. TGF-beta isoforms in renal fibrogenesis. Kidney Int 2003; 64: 844–56

48 Khan SR, Glenton PA, Byer KJ. Modeling of hyperoxaluric calcium oxalate nephrolithiasis: experimental induction of hyperoxaluria by hydroxy-L-proline. Kidney Int 2006; 70: 914–23

Correspondence: Piyaratana Tosukhowong and Chanchai Boonla, Department of Biochemistry, Faculty of Medicine, Chulalongkorn University, Bangkok, 10330 Thailand.e-mail: piyaratana_t@yahoo.com; chanchai.b@chula.ac.th

Abbreviations: MCP-1, monocyte-chemoattractant protein-1; IL-6, interleukin6; EMT, epithelial-mesenchymal transition; ααααSMA,, α-smooth muscle actin; Cr, creatinine; CCr, and Cr clearance; H&E, haematoxylin and eosin; IQR, interquartile range; BMI, body mass index; CaOx, calcium oxalate; CaP, calcium phosphate; UA, uric acid.