Báo cáo khoa học: Expression of cholinesterases in human kidney and its variation in renal cell carcinoma types pdf

To fill the gap and to determine whether cholinesterases are abnormally expressed in renal tumours, paired pieces of normal kidney and renal cell carcinomas RCCs were compared for choline

variation in renal cell carcinoma types

Encarnacio´n Mun˜oz-Delgado1, Marı´a Fernanda Montenegro1, Francisco Javier Campoy1,

Marı´a Teresa Moral-Naranjo1, Juan Cabezas-Herrera2, Gyula Kovacs3and Cecilio J Vidal1

1 Department of Biochemistry and Molecular Biology-A, University of Murcia, Spain

2 Research Unit of Clinical Analysis Service, University Hospital Virgen de la Arrixaca, Murcia, Spain

3 Laboratory of Molecular Oncology, Medical Faculty, Ruprecht-Karls-University of Heidelberg, Germany

Keywords

chromophobe renal cell carcinoma (chRCC);

conventional renal cell carcinoma (cRCC);

glycosylphosphatidylinositol anchor; papillary

renal cell carcinoma (pRCC); renal

oncocytoma

Correspondence

C J Vidal, Department of Biochemistry and

Molecular Biology-A, University of Murcia,

Campus de Espinardo, E-30071 Murcia,

Spain

Fax: +34 868 884147

Tel: +34 868 884774

E-mail: cevidal@um.es

(Received 9 July 2010, revised 10 August

2010, accepted 3 September 2010)

doi:10.1111/j.1742-4658.2010.07861.x

Despite the aberrant expression of cholinesterases in tumours, the question of their possible contribution to tumorigenesis remains unsolved The identifica-tion in kidney of a cholinergic system has paved the way to funcidentifica-tional studies, but details on renal cholinesterases are still lacking To fill the gap and to determine whether cholinesterases are abnormally expressed in renal tumours, paired pieces of normal kidney and renal cell carcinomas (RCCs) were compared for cholinesterase activity and mRNA levels In studies with papillary RCC (pRCC), conventional RCC, chromophobe RCC, and renal oncocytoma, acetylcholinesterase activity increased in pRCC (3.92 ± 3.01 mUÆmg)1, P = 0.031) and conventional RCC (2.64 ± 1.49 mUÆmg)1,

P= 0.047) with respect to their controls (1.52 ± 0.92 and 1.57 ± 0.44 mUÆmg)1) Butyrylcholinesterase activity increased in pRCC (5.12 ± 2.61 versus 2.73 ± 1.15 mUÆmg)1, P = 0.031) Glycosylphosphatidylinositol-linked acetylcholinesterase dimers and hydrophilic butyrylcholinesterase tetramers predominated in control and cancerous kidney Acetylcholinesterase mRNAs with exons E1c and E1e, 3¢-alternative T, H and R acetylcholinesterase mRNAs and butyrylcholinesterase mRNA were identified in kidney The levels

of acetylcholinesterase and butyrylcholinesterase mRNAs were nearly 1000-fold lower in human kidney than in colon Whereas kidney and renal tumours showed comparable levels of acetylcholinesterase mRNA, the content of butyrylcholinesterase mRNA was increased 10-fold in pRCC The presence

of acetylcholinesterase and butyrylcholinesterase mRNAs in kidney supports their synthesis in the organ itself, and the prevalence of glycosyl-phosphatidylinositol-anchored acetylcholinesterase explains the splicing to acetylcholinesterase-H mRNA The consequences of butyrylcholinesterase upregulation for pRCC growth are discussed

Structured digital abstract

l MINT-7992181 : BuChE (uniprotkb: P06276 ) and BuChE (uniprotkb: P06276 ) bind ( MI:0407 )

by chromatography technology ( MI:0091 )

l MINT-7992175 : AChE (uniprotkb: P22303 ) and AChE (uniprotkb: P22303 ) bind ( MI:0407 ) by chromatography technology ( MI:0091 )

Abbreviations

ACh, acetylcholine; AU, arbitrary units; Brij 96, polyoxyethylene-oleyl ether; chRCC, chromophobe renal cell carcinoma; cRCC, conventional renal cell carcinoma; GPI, glycosylphosphatidylinositol; Iso-OMPA, tetraisopropyl pyrophosphoramide; LCA, Lens culinaris agglutinin; LOH, loss of heterozygosity; NK, normal kidney; PIPLC, phosphatidylinositol-specific phospholipase C; pRCC, papillary renal cell carcinoma; RCA, Ricinus communis agglutinin; RCC, renal cell carcinoma; RO, renal oncocytoma.

Renal cell carcinomas (RCCs), which account for

2% of adult malignancies, affect about 150 000

peo-ple per year worldwide [1], more than 85 000 in

European countries and nearly 3000 in Spain, with

1647 deaths in 2005 (WHO-IARC http://www-dep

iarc.fr/) RCCs arise from the renal tubular

epithe-lium, and, according to their morphological and

molecular criteria, are classified into malignant and

more indolent parenchymal neoplasms [2,3] The

World Health Organization classification distinguishes

clear cell (conventional) RCC (cRCC) (70–80%),

papillary (chromophile) RCC (pRCC) (10–15%),

chromophobe RCC (chRCC) (3–5%), tumours of

collecting ducts of Bellini (1%), and Xp11

transloca-tion RCC and unclassified RCC (together 1%)

Benign or less severe tumours are papillary renal cell

adenoma, metanephric adenoma, and renal

oncocy-toma (RO) Deletions at chromosome 3p, mutations

of the VHL suppressor gene and loss of

heterozygos-ity (LOH) of 9p, 14q, 17p and 10q have been

reported in cRCC [4] Trisomy 7 and 17 and loss of

the Y-chromosome are frequent in pRCC [5] Loss

at 9p13 has been associated with poor survival in

patients with pRCC [4], and, despite the good

prog-nosis of chRCC, loss of various chromosomes,

muta-tions of p53, LOH at 10q23.3 and telomere

shortening have been reported [6] RO consists of

mixtures of cells with normal and abnormal

karyo-types [7] About 4% of renal cancers arise from

hereditary syndromes [8]

Acetylcholinesterase (UniProt P22303) and

butyr-ylcholinesterase (P06276) are enzymes that rapidly

hydrolyse acetylcholine (ACh) The human

acetyl-cholinesterase gene maps at 7q22 [9] and the

butyryl-cholinesterase gene at 3q26 [10] A range of

3¢-alternatively spliced and 5¢-alternatively spliced

ace-tylcholinesterase mRNAs have been identified [11,12]

The 3¢-alternative mRNAs code for the three classical

catalytic acetylcholinesterase subunits: ‘tailed’ or

‘syn-aptic’ (T, P22303-1), ‘hydrophobic’ or ‘erythrocytic’

(H, P22303-2), and ‘readthrough’ (R, P22303-4) [11,13]

Acetylcholinesterase-T forms homo-oligomers, the

so-called ‘globular forms’ (G1, G2, and G4), and

het-ero-oligomers, depending on the lack or addition of

structural subunits Acetylcholinesterase-H adds

glyco-sylphosphatidylinositol (GPI) and forms amphiphilic

monomers (G1A) and dimers (G2A) [14] The

stress-inducible acetylcholinesterase-R subunit can replace the

T-subunit in oligomers [11] Five 5¢-alternative

acetyl-cholinesterase mRNAs have been identified in mice and

three in humans [11,12], and acetylcholinesterase-H,

acetylcholinesterase-T and acetylcholinesterase-R mRNAs starting with exon E1e code for N-terminally extended acetylcholinesterase subunits (N-acetylcholin-esterase), whose extension displays a membrane anchorage motif [11] A single butyrylcholinesterase mRNA exists, and its protein product forms G1, G2 and G4species, with hydrophilic or amphiphilic proper-ties according to the folding of the butyrylcholinester-ase subunit [15]

Besides their hydrolytic action, both acetylcholines-terase and butyrylcholinesacetylcholines-terase seem to play noncata-lytic roles, which would explain their wide distribution

in tissues and cells [16], including stem cells [17] Increasing evidence links these noncatalytic actions with the binding of cholinesterases to several protein partners Thus, it has been reported that acetylcholin-esterase can interact with laminin and collagen IV [17,18], nicotinic receptors [19], amyloid b-peptide and presenilin-1 [20], neuronal enolase, the scaffold protein RACK1 and protein kinase C [21], the corepressor CtBP [22], and Ran-binding protein [23] Butyrylcho-linesterase can also have noncatalytic actions, as judged by the role of the butyrylcholinesterase-K–apo-lipoprotein Ee4–amyloid b-peptide complex in Alzhei-mer’s disease [24,25] and of butyrylcholinesterase itself

in megakaryocytopoiesis suppression and retinal cell differentiation [16]

The expression of acetylcholinesterase and butyr-ylcholinesterase in neural and non-neural tumours [26] and the amplification of their genes in leukae-mias and ovarian cancer [16,26] support a role for cholinesterases in carcinogenesis This notion is given weight by the aberrant expression and structural changes of acetylcholinesterase and butyrylcholinest-erase in cancers of diverse origin [26,27], the tumour-inducing effect of anticholinesterase agents [26,28], the relationship between astrocytoma severity and acetylcholinesterase expression [27], the role of acetyl-cholinesterase in apoptosis [23], and the downregula-tion of cholinesterases in metastasized lymph nodes [29], as well as in colorectal [30] and lung [31] can-cers

Despite the long time that has elapsed since the observation of cholinesterases in mouse kidney [32] and MDCK cells [33], and, more recently, of ACh and cholinergic receptors in the human urothelium [34], the expression of cholinesterases in human kidney has not been studied yet The present research was intended to partially fill this gap by examining the levels of cholin-esterase activity and mRNAs in human kidney and their possible variation in tumours

Cholinesterase activity in human kidney and

renal tumours

Butyrylcholinesterase activity predominated over

ace-tylcholinesterase activity in healthy kidney (Table 1)

As compared with control specimens,

acetylcholinester-ase activity was 2.6-fold and 1.7-fold increacetylcholinester-ased in

pRCC and cRCC (P < 0.05) (Table 1)

Acetylcholin-esterase activity also rose in chRCC and RO, but in a

non-statistically significant manner

Butyrylcholinester-ase activity increButyrylcholinester-ased 1.9-fold in pRCC (P < 0.05)

Most acetylcholinesterase (77 ± 20%) and

butyrylcho-linesterase activities (92 ± 10%) in unaffected and

cancerous kidney were released by the two-step

extrac-tion procedure Acetylcholinesterase activity was recov-ered to a lower extent in the S1 supernatant (17 ± 12%) than in the S2 supernatant (60 ± 15%), unlike butyrylcholinesterase activity (70 ± 12% in S1 and 22 ± 7% in S2) No statistically significant differ-ences between unaffected and cancerous pieces were observed in the extent of cholinesterase extraction

Molecular species of acetylcholinesterase in unaffected and cancerous kidney

Sedimentation analysis with normal kidney (NK) extracts revealed principal acetylcholinesterase species with a sedimentation coefficient of 4.0 ± 0.3S (80 ± 7%) and less abundant species with a sedimenta-tion coefficient of 2.4 ± 0.3S (20 ± 5%) (Fig 1A)

Table 1 Acetylcholinesterase and butyrylcholinesterase activities in noncancerous kidney (Control) and renal cell carcinoma (Tumour) Activi-ties are given as mean ± standard deviation; 1 mU of cholinesterase activity is equal to 1 nmol of substrate split per minute P-values were calculated with the Wilcoxon signed rank test; bold type indicates significant differences for 95% confidence.

Acetylcholinesterase (mU per mg protein) Butyrylcholinesterase (mU per mg protein)

Fig 1 Distribution of cholinesterase species in human kidney and RCCs (A) Representative sedimentation profiles with acetylcholinesterase species in S1 + S2 supernatants of NK and pRCC (B) Cleavage of the hydrophobic moiety in renal acetylcholinesterase Sedimentation patterns showing acetylcholinesterase species in samples incubated without (PIPLC–) and with hydroxylamine and PIPLC (PIPLC+); see Experimental procedures (C) Sedimentation patterns with butyrylcholinesterase species in S1 + S2 supernatants of NK and pRCC C and P

in profiles denote catalase and alkaline phosphatase.

According to their sedimentation coefficients and

phe-nyl-agarose adsorption (Fig 2, sedimentation profile for

fraction F82), the 4.0S and 2.4S forms were assigned to

amphiphilic dimers (G2A) and monomers (G1A) [35]

Their synthesis was unaffected by malignancy, as judged

by the almost unchanged distribution of

acetylcholines-terase species in NK, pRCC (Fig 1A), cRCC, chRCC

and RO (sedimentation profiles not shown)

The observation of GPI-anchored

acetylcholinester-ase in epithelial tissues [36,37] prompted us to study

the sensitivity of renal acetylcholinesterase to

phospha-tidylinositol-specific phospholipase C (PIPLC) The

results showed that, in contrast to the amphiphilic

GPI-linked G2A acetylcholinesterase of beef

erythro-cytes, which became fully hydrophilic (G2H) after

treat-ment with phospholipase, only 20% of the renal G2A

species did so (profiles not shown) Nevertheless, the

fact that the conversion extent was increased to 40–

50% by prior incubation with hydroxylamine (Fig 1B)

revealed GPI moieties in at least half of renal

acetyl-cholinesterase Although the low sensitivity of kidney

(Fig 1B) and erythrocyte acetylcholinesterase to

PIP-LC [38] might suggest a blood origin for renal

acetyl-cholinesterase, the presence in kidney of G1A

acetylcholinesterase, which is absent from erythrocytes,

and the difference between acetylcholinesterases of

kid-ney and erythrocytes in the extent of binding with the

lectins concanavalin A, Lens culinaris agglutinin

(LCA), and Ricinus communis agglutinin (RCA) (Fig 3), ruled out the blood origin and supported the renal cells themselves as the most probable source of kidney acetylcholinesterase

Butyrylcholinesterase species in healthy kidney and RCC

The kidney butyrylcholinesterase activity distributed between principal 12.1 ± 0.2S species (70 ± 7%) and less abundant 4.9 ± 0.2S species (30 ± 12%) (Fig 1C) According to their sedimentation coefficients and hydrophilic properties, as judged by their inability

to be retained in phenyl–agarose (Fig 2, profiles for fractions F6–F10), the butyrylcholinesterase species were assigned to hydrophilic tetramers (G4H) and monomers (G1H) It is worth noting the profitable use

of phenyl–agarose to resolve not only hydrophilic and amphiphilic cholinesterase species [37] but also hydro-philic butyrylcholinesterase tetramers and monomers, taking advantage of the faster elution of the former (Fig 2, profiles F6–F10) Although the G4H butyrylch-olinesterase species were always more abundant than the G1H species (Fig 1C), subtle differences between normal samples and cancerous pieces (even from the same tumour type) in the proportion of butyrylcholin-esterase species prevented us from ascertaining possible changes in their distribution as the result of cancer

Fig 2 Phenyl–agarose chromatography with renal cholinesterases The S1 superna-tant of NK was passed through phenyl– agarose, and fractions with unbound and Triton X-100 (TX-100)-eluted cholinesterase activity were assayed for acetylcholinester-ase and butyrylcholinesteracetylcholinester-ase Cholinester-ase species in butyrylcholinesterCholinester-ase-rich fractions (F6–F10) and in the acetylcholines-terase-rich fraction (F82) were identified by centrifugation as in Fig 1.

Levels of acetylcholinesterase and

butyrylcholinesterase mRNAs

Regardless of healthy or pathological status, human

kidney contained acetylcholinesterase mRNAs with

exons E1c and E1e, the three 3¢-alternatively spliced

acetylcholinesterase mRNAs (R, H, and T), and the

butyrylcholinesterase transcript (Fig 4) Although

RT-PCR quantifications are not completely reliable,

and only give an approximate idea of the relative

content of mRNAs, real-time PCR results allowed us

to detect low levels of acetylcholinesterase mRNAs

in kidney Thus, unaffected renal pieces displayed

comparable quantities of acetylcholinesterase mRNAs

with E1c (96 ± 62 copies per 106 copies of b-actin

mRNA) and E1e (148 ± 80 copies) The E1a-bearing

acetylcholinesterase mRNA was undetected in renal pieces No significant differences between unaffected kidney, pRCC, cRCC, chRCC and RO in the con-tent of the 5¢-alternative acetylcholinesterase mRNAs were observed

Concerning the 3¢-alternative acetylcholinesterase mRNAs, NK had similar amounts of acetylcholinester-ase-R (30 ± 17 copies) and acetylcholinesterase-H (24 ± 19 copies) mRNAs, and their quantities were unmodified in the different classes of tumours The amount of acetylcholinesterase-T mRNA in control kidney (81 ± 67 copies) did not statistically vary in pRCC, cRCC, and RO, and tended to decrease in chRCC (20 ± 10 copies; P = 0.06) (Fig 4) Finally, the level of butyrylcholinesterase mRNA in unaffected kidney (19 ± 12 copies) was little changed in cRCC, chRCC, and RO, but significantly increased in pRCC (237 ± 161 copies, P = 0.008) (Fig 4)

Discussion The histochemical observation of acetylcholinesterase and butyrylcholinesterase in mammalian kidney [32] and canine MDCK renal cells [33] justified a detailed biochemical study of renal cholinesterases, but it had not yet been performed The present results and our previous data showing GPI-anchored acetylcholinester-ase dimers and monomers in human kidney (Figs 1 and F2, F82), meningioma [37], breast [36], lung [31] and gut [30] demonstrate the capacity of epithelial tis-sues for translating the acetylcholinesterase-H mRNA, and undermines the widely accepted idea that the GPI-bound acetylcholinesterase of mammals arises almost exclusively from blood cells

As in human gut [30], butyrylcholinesterase activity prevailed over acetylcholinesterase activity in kidney The comparable distribution of G4H and G1H

Fig 4 Real-time PCR results showing acetylcholinesterase and

butyrylcholinesterase mRNA levels in control kidney and renal

tumours Mean values of five or six determinations with six or

seven paired samples of unaffected and cancerous kidney.

*P < 0.001.

Fig 3 Lectin interaction patterns of

cho-linesterases of human kidney, erythrocytes,

and blood plasma Extracts of kidney and

erythrocytes, along with blood plasma

sam-ples, were incubated with lectin-free

Sepha-rose 4B (control) and SephaSepha-rose-linked

lectins Then, the agarose beads with bound

cholinesterase activity were removed, and

the unbound cholinesterase activity was

assayed The percentage of lectin-bound

activity was calculated by comparing

cholin-esterase activity in lectin-incubated and

control assays Results are means of four

experiments *P < 0.05, **P < 0.01.

butyrylcholinesterase in kidney (Fig 1C), meningioma

[37], breast [36], and colon [30], and the observation

of butyrylcholinesterase mRNA in kidney (Fig 4),

colon [30], and cancerous cell lines [39], support the

idea of butyrylcholinesterase synthesis in the organs

themselves Nevertheless, the great quantity of

butyr-ylcholinesterase activity in blood plasma [40] might

lead us to think that renal butyrylcholinesterase

arises totally or in part from blood However,

bear-ing in mind the need for vigorous irrigation to

favour tumour growth and the abundance of G4H

butyrylcholinesterase in plasma, if blood were the

source of kidney butyrylcholinesterase, an appreciable

increase in butyrylcholinesterase activity of chRCC,

cRCC, and RO, instead of its invariability (Table 1),

should have been expected, along with a robust

increase in the proportion of G4H

butyrylcholinester-ase The incomplete binding of renal

butyrylcholinest-erase to the lectin LCA, which fully reacts with the

enzyme of plasma (Fig 3), strongly supports the

renal origin of the butyrylcholinesterase activity

assayed in kidney This proposal is in agreement

with the cytochemical staining of acetylcholinesterase

in the capsule of Bowman [41] and of

acetylcholines-terase and butyrylcholinesacetylcholines-terase in the glomerulus

and the tubule, which is stronger for

acetylcholines-terase in the rough endoplasmic reticulum of

mesan-gial cells and for butyrylcholinesterase in the

reticulum of endothelial cells [32] Thus, it is likely

that epithelial, mesangial and endothelial cells can all

contribute to renal cholinesterase activity, but the

most important point is the comparable profiles of

cholinesterase forms in control and cancerous kidney,

which rules out major changes in their biosynthesis

as the result of cancer

As regards the range of cholinesterase mRNAs,

kid-ney and renal tumours share the capacity to express

the three 3¢-spliced mRNAs (R, H, and T) and the

5¢-spliced acetylcholinesterase mRNAs that start with

E1c and E1e (Fig 4) As the E1e acetylcholinesterase

mRNA codes for N-terminally extended

acetylcholin-esterase [11], whose extension is selectively associated

with apoptosis of neural cells [42], the E1e mRNA

might behave as a brake to prevent or attenuate tumour

progression in kidney and other organs As expected,

human kidney contained much less acetylcholinesterase

mRNA ( 150 copies for acetylcholinesterase-R +

acetylcholinesterase-H + acetylcholinesterase-T mRNAs)

(Fig 4) than gut ( 2500 copies) [30], mouse brain

( 35 000 copies) [12], or muscle ( 10 000 copies)

[43] The presence of acetylcholinesterase-T mRNA in

kidney (Fig 4) contrasts with the absence of catalytic

acetylcholinesterase-T protein from kidney and

cancer-ous cell lines of lung, breast, and gut [39], a feature that might be attributed to microRNA-induced trans-lational repression of the acetylcholinesterase-T mRNA in epithelial cells In this respect, there is evidence of a regulatory role for microRNA-132 in the expression of acetylcholinesterase in leukocytes [44], but other reasons may exist, e.g fast degradation of acetylcholinesterase-T protein, rapid secretion of oligomers [13], or synthesis of catalytically incompetent protein [14]

Concerning the variation of cholinesterase activity in renal tumours, the 2.6-fold and 1.7-fold increased acetylcholinesterase activities in pRCC and cRCC (Table 1), despite their unchanging levels of acetylcho-linesterase mRNAs (Fig 4), point to malignancy-driven changes in translational efficiency This increase

of acetylcholinesterase activity in pRCC and cRCC is

in agreement with the upregulation of acetylcholines-terase in tumour cell lines [27], but not with the reduc-tion of activity in cancerous lymph nodes [29] and gut [30] The reports in ovarian carcinoma showing ampli-fication of the ACHE gene on the one hand, and frequent LOH at 7q22 on the other [26], besides the finding of a negative correlation between upregulation

of acetylcholinesterase with androgens in ovarian cancer and patient survival [45], illustrate how complex the changes in acetylcholinesterase expression can

be in ovarian tumours and, most probably, in other cancers [27]

Our observations regarding the higher acetylch-olinesterase activity in kidney tumours suggest that a relationship may exist between increased acetylcholin-esterase activity and cell proliferation, a link that has also been suggested for hyperproliferation of lympho-cytes in thymomas [46] However, other studies have implicated acetylcholinesterase in apoptotic cell death [47] These ideas are not necessarily contradictory, if one considers the widely accepted idea that the effects

of acetylcholinesterase depend on the cell type and differentiation state, the levels of 5¢-spliced and 3¢-spliced acetylcholinesterase mRNAs, their lifespan and translational efficiency, and the capacity of the translated product to bind to protein partners

The upregulation of butyrylcholinesterase in pRCC (Table 1 and Fig 4) is in agreement with previous reports on squamous cell lung [48], breast [49] and hepatic [50] carcinomas In this respect, it is worth mentioning the recommended use of butyrylcholinest-erase overexpression as a predictive survival index in liver cancer [51] Considering the butyrylcholinesterase contribution to immortalization of several SV40-trans-formed cell types and to maturation of megakaryo-cytes [16], a role for butyrylcholinesterase in the

proliferation⁄ differentiation of different cell types has

been proposed, and although some information on this

issue is available for butyrylcholinesterase-null retinal

cells [52], further research is required to assess the

involvement of butyrylcholinesterase in cell

prolifera-tion and cancer The higher increase in

butyrylcholin-esterase mRNA than activity levels in pRCC may

point at tumour-related elevations in

butyrylcholinest-erase-targeted micro-RNA(s) In contrast,

acetylcholin-esterase mRNA levels remained unchanged and the

enzymatic activity increased in pRCC This may

inver-sely reflect a tumour-associated decline in

acetylcholin-esterase mRNA-targeted micro-RNA(s) Given the

increasing importance of micro-RNAs in tumorigenic

processes, the proposal should be seriously considered

in further studies

The increased acetylcholinesterase and

butyrylcho-linesterase activities in pRCC (Table 1) may indeed

represent a side effect of the transformed cell

pheno-type, but the binding of the cytotoxic cisplatin to

acetylcholinesterase [53] and probably to

butyrylcho-linesterase suggests a relationship between the

increased cholinesterase activity and pRCC

chemore-sistance Nevertheless, the crucial question is whether

the increase in cholinesterase activity contributes or

not to pRCC growth Obviously, any increase in

cho-linesterase activity would reduce the availability of

ACh and would therefore impair cholinergic

responses The origin of pRCC in the tubular

epithe-lium [54] and the presence in it of muscarinic and

nicotinic receptors [34,55] support a role for ACh in

renal tubules The increase in cholinesterase activity

in pRCC, the subsequent decrease in ACh availability

and the lowered cholinergic activation may have

pathological consequences, but further research is

needed to clarify functional aspects of cholinergic

sig-nalling in the urothelium

In summary, our results show that human kidney

contains abundant butyrylcholinesterase activity,

distributed among G4H and G1H species, and less

acetylcholinesterase activity as GPI-anchored species

Whereas the observation of acetylcholinesterase and

butyrylcholinesterase mRNAs in kidney supports their

synthesis in the organ itself, the prevalence of

GPI-linked acetylcholinesterase in kidney and other

epithe-lial tissues explains their acetylcholinesterase-H mRNA

content The fact that G4Hand G1H

butyrylcholinester-ase are similarly distributed in various epithelia

sup-ports their programmed synthesis The overexpression

of cholinesterases in pRCC contrasts with their

under-expression in cancerous lymph nodes and gut, and

these features highlight the complex regulation of

cho-linesterases in cancer

Experimental procedures Materials

Acetylthiocholine and butyrylthiocholine iodide, 5,5¢-di-thiobis(2-nitrobenzoic acid), 1,5-bis(4-allyldimethylam-moniumphenyl)-pentan-3-one dibromide (BW284c51), tetraisopropyl pyrophosphoramide (Iso-OMPA), Brij 96, antiproteinases, protein markers for sedimentation analysis (beef liver catalase and bovine intestine alkaline phospha-tase), phenyl–agarose, lectin-free Sepharose 4B and agarose-bound concanavalin A, LCA, RCA, DNase I, ethidium bromide and DNA size markers were all purchased from Sigma (St Louis, MO, USA) Moloney murine leukaemia virus reverse transcriptase, random primers and the Purelink Micro-to Midi total RNA Purification System for total RNA extraction were provided by Invitrogen (Carlsbad,

CA, USA), and dNTPs by Eppendorf (Hamburg, Germany) TaqMan PCR Master Mix was from Applied Biosystems (Foster City, CA, USA), and ribonuclease inhibitor from Amersham-Pharmacia (Buckinghamshire, UK) PIPLC of Bacillus thuringiensiswas kindly donated by N M Hooper (University of Leeds, UK)

Patients and tumours Kidney specimens were taken from patients who had under-gone tumour nephrectomy They were properly informed about the use of samples for research After surgery, paired samples of renal tumours and adjacent unaffected tissue were taken, snap-frozen in liquid nitrogen, and stored at )80 C Histological diagnosis was made according to the Heidelberg Classification of Renal Cell Tumours [2] Seven specimens of pRCC, seven of cRCC, six of chRCC, and six

of RO, besides adjacent pieces of unaffected tissue, were used in this study This research was approved by the ethics committee of the University of Murcia

Extraction and assay of cholinesterases Kidney pieces were homogenized with detergent-free NaCl⁄ Tris (1 m NaCl, 50 mm MgCl2, 10 mm Tris, pH 7.0) containing the antiproteinases benzamidine (2 mm), pepsta-tin A (10 lgÆmL)1), leupeptin (20 lgÆmL)1), aprotinin (20 UÆmL)1), soybean trypsin inhibitor (0.1 mgÆmL)1), and bacitracin (1 mgÆmL)1) After centrifugation at 170 000 g for 1 h at 4C in a 70 Ti rotor (Beckman, Palo Alto, CA, USA), the S1 supernatant with loosely bound

cholinesteras-es was saved The pellet was re-extracted with NaCl⁄ Tris supplemented with 1% Brij 96 and antiproteinases After centrifugation as above, the S2 supernatant with tightly bound cholinesterases was recovered Acetylcholinesterase was extracted from human erythrocytes as reported elsewhere [38]

Cholinesterase activity was determined by the Ellman

method: acetylcholinesterase with 1 mm acetylthiocholine

and 50 lm Iso-OMPA, and butyrylcholinesterase with

1 mm butyrylthiocholine and 10 lm BW284c51 [36]

Unspe-cific esterase activity, measured in assays including both

BW284c51 and Iso-OMPA, was discounted for the

calcula-tion of true acetylcholinesterase and butyrylcholinesterase

activities Cholinesterase activity is given in nanomoles of

the preferred substrate hydrolysed per min at 25C (mU)

Acetylthiocholine hydrolysis attributable to unspecific

esterases in unaffected and cancerous pieces amounted to

15–25%, and that of butyrylthiocholine to 20–30% True

cholinesterase activity was calculated by subtracting the

unspecific hydrolysis from the total hydrolysis of the

substrate Cholinesterase activity in sedimentation profiles

is given in arbitrary units (AU), in which case one unit of

activity refers to an increase of 0.001 absorbance units per

microlitre of sample and per min, but normalized for the

volume of sample added to the gradient Protein

concentra-tion was determined by the Lowry method [36]

Characterization of cholinesterase components

Acetylcholinesterase and butyrylcholinesterase species were

resolved by sedimentation analysis and identified by their

sedimentation coefficients [36] A mixture of the S1 and S2

supernatants (0.5 mL each) plus the sedimentation markers beef liver catalase (11.4S) and intestine alkaline phospha-tase (6.1S) was loaded onto 5–20% sucrose gradients con-taining 0.5% Brij 96 The gradient tubes were centrifuged

at 170 000 g for 24 h at 4C in a Beckman SW41Ti rotor Overlapping peaks were resolved with the peak-fit soft-ware from SPSS The percentage of each cholinesterase form was determined by comparing cholinesterase activity under each peak area and under the entire profile

Amphiphilic acetylcholinesterase and hydrophilic butyr-ylcholinesterase in the S1 supernatant of kidney were separated by taking advantage of the capacity of phenyl– agarose to adsorb amphiphilic cholinesterases [37] In addition, the faster elution of butyrylcholinesterase tetramers permitted their separation from butyrylcholinest-erase monomers [12] The presence of GPI residues in renal acetylcholinesterase was tested by its exposure to PIPLC of

B thuringiensis Samples were incubated in the absence (control) and presence of PIPLC, both without and with prior treatment with alkaline hydroxylamine, as reported previously [30]

Lectin interaction assays allowed us to distinguish between homologous cholinesterase forms of kidney and blood Mixtures of S1 and S2 extracts of kidney were incu-bated with lectin-free Sepharose 4B (control) and with Sepharose-linked lectins Samples of Triton X-100-extracted

A

B

Fig 5 Primers used to quantify acetylcho-linesterase and butyrylchoacetylcho-linesterase mRNAs by real-time PCR (A) Scheme showing the position of the primers (B) Pri-mer sequences and PCR product sizes Gene ID and accession numbers are as fol-lows: ACHE, 43 and ENSG 00000087085; BCHE, 590 and ENSG 0000011420; and ACTB (b-actin), 60 and ENSG 00000075624.

acetylcholinesterase from human erythrocytes and from

blood plasma were also incubated After incubation,

the lectin–cholinesterase complexes were removed by

centrifugation at 3000 g for 5 min at 4C in microcentrifuge

(Denver Instrument Company, Argada, CO, USA), and the

unbound cholinesterase activity was assayed The percentage

of lectin-bound activity was determined by comparing

the activity in lectin-incubated and control assays [12]

Quantification of cholinesterase mRNAs by real

time RT-PCR

Total RNA was extracted from frozen renal specimens with

the Purelink Micro-to Midi total RNA Purification System,

after a first extraction with Trizol For reverse

transcrip-tion, 5 lg of DNAse I-treated RNA was denatured at

70C for 10 min and cooled rapidly A mixture of buffer,

dithiothreitol, dNTPs, random primers and ribonuclease

inhibitor was added before heating for 2 min at 42C

Then, 200 U of Moloney murine leukaemia virus reverse

transcriptase was added, and synthesis of cDNAs was

per-formed for 50 min at 42C in a volume of 20 lL Finally,

samples were heated for 15 min at 72C and kept frozen

For PCR, primer pairs were designed to amplify the

cDNAs derived from the 5¢-alternative acetylcholinesterase

mRNAs that include exons E1c and E1e, the 3¢-alternative

acetylcholinesterase mRNAs (R, H, or T), the

butyrylcho-linesterase mRNA, and the b-actin mRNA, used as the

internal standard The sequence and position of the

prim-ers, as well as the size of the PCR products, are provided

in Fig 5 cDNA was amplified in an Applied

Biosys-tems 7500 real-time PCR system, using a MicroAmp

Opti-cal 96-well Reaction Plate with 25 lL of reaction volume

The buffered medium contained 2 lL of variable dilutions

of cDNA, 0.2 lm specific primers, and the TaqMan PCR

master mix Reactions comprised a first step of 10 min at

95C, followed by 45 cycles of 15 s at 95 C and 60 s at

60C A final dissociation stage allowed us to study the

melting curves The relative contents of acetylcholinesterase

and butyrylcholinesterase cDNAs, with respect to b-actin

cDNA, was determined by the 2)DCt method PCR

pro-ducts were separated in 3% agarose gels and visualized

with ethidium bromide Their lengths, calculated with DNA

size markers and gelpro software, coincided with the

expected sizes For reliability, the PCR products derived

from acetylcholinesterase-R, acetylcholinesterase-E1c and

acetylcholinesterase-E1e mRNAs were sequenced in a

Genetic Analyzer ABI Prism 3130 (Applied Biosystems)

The relative amounts of cholinesterase mRNAs are given as

number of copies per 106copies of the b-actin mRNA

Statistical analysis

The results are expressed as mean ± standard deviation

Statistical differences in cholinesterase activity between

nor-mal and nor-malignant kidney pieces were assessed with the Wilcoxon signed rank test Data were analysed by consider-ing paired samples (control and neoplastic samples of the same patient) The significance of differences in lectin bind-ing to cholinesterases was evaluated with Student’s t-test

Acknowledgements

We thank N Hooper (University of Leeds, UK) for providing us with PIPLC from B thuringiensis and Centro Nacional de Investigaciones Oncolo´gicas of Spain (CNIO), as well as J E Herna´ndez-Barcelo´ and

F Ruiz-Espejo (Hospital Virgen de la Arrixaca of Murcia, Spain) for the kind donation of unaffected kidney, renal cancer and blood samples This research was supported by the Instituto de Salud Carlos III of Spain (Grant FIS-PI041504) and the Fundacio´n

Se´ne-ca of Murcia (Grant 08648⁄ PI08), which also provided

a scholarship for M F Montenegro

References

1 Jemal A, Siegel R, Ward E, Hao X, Xu J & Thun MJ (2009) Cancer statistics CA Cancer J Clin 59, 225–249

2 Kovacs G, Akhtar M, Beckwith BJ, Bugert P, Cooper

CS, Delahunt B, Eble JN, Fleming S, Ljungberg B, Medeiros LJ et al (1997) The Heidelberg classification

of renal cell tumours J Pathol 183, 131–133

3 Moch H, Gasser T, Amin MB, Torhorst J, Sauter G & Mihatsch MJ (2000) Prognostic utility of the recently recommended histologic classification and revised TNM staging system of renal cell carcinoma A Swiss experi-ence with 588 tumors Cancer 89, 604–614

4 Hansel DE (2006) Genetic alterations and histopatho-logic findings in familial renal carcinoma Histol Histo-pathol 21, 437–444

5 Kovacs G, Fuzesi L, Emanual A & Kung HF (1991) Cytogenetics of papillary renal cell tumors Genes Chromosomes Cancer 3, 249–255

6 Brunelli M, Eble JN, Zhang S, Martignoni G, Delahunt B & Cheng L (2004) Eosinophilic and classic chromophobe renal cell carcinomas have similar frequent losses of multiple chromosomes from among chromosomes 1, 2, 6, 10, and 17, and this pattern of genetic abnormality is not present in renal oncocytoma Mod Pathol 18, 161–169

7 Zubakov D, Stupar Z & Kovacs G (2006) Differential expression of a new isoform of DLG2 in renal onco-cytoma BMC Cancer 6, 106 doi:10.1186/1471-2407-6-106

8 Linehan WM, Bratslavsky G, Pinto PA, Schmidt LS, Neckers L, Bottaro DP & Srinivasan R (2010) Molecu-lar diagnosis and therapy of kidney cancer Annu Rev Med 61, 329–343

9 Getman DK, Eubanks JH, Camp S, Evans GA &

Taylor P (1992) The human gene encoding AChE is

located on the long arm of chromosome 7 Am J Hum

Genet 51, 170–177

10 Allderdice PW, Gardner HA, Galutira D, Lockridge O,

La Du BN & McAlpine PJ (1991) The cloned

butyr-ylcholinesterase (BCHE) gene maps to a single

chromo-some site, 3q26 Genomics 11, 452–454

11 Meshorer E & Soreq H (2006) Virtues and woes of

AChE alternative splicing in stress-related

neuropathol-ogies Trends Neurosci 29, 216–224

12 Ferna´ndez-Go´mez FJ, Mun˜oz-Delgado E,

Montenegro MF, Campoy FJ, Vidal CJ & Jorda´n J

(2010) Cholinesterase activity in brain of

senescence-accelerated-resistant mouse SAMR1 and its variation in

brain of senescence-accelerated-prone mouse SAMP8

J Neurosci Res 88, 155–166

13 Massoulie´ J, Perrier N, Noureddine H, Liang D &

Bon S (2008) Old and new questions about

cholinesterases Chem Biol Interact 175, 30–44

14 Nieto-Cero´n S, Sa´nchez del Campo LF,

Mun˜oz-Delgado E, Vidal CJ & Campoy FJ (2005) Muscular

dystrophy by merosin deficiency decreases

acetylcholinesterase activity in thymus of Lama2dy

mice J Neurochem 95, 1035–1046

15 Altamirano CV & Lockridge O (1999) Conserved

aro-matic residues of the C-terminus of human

butyrylcho-linesterase mediate the association of tetramers

Biochemistry 38, 13414–13422

16 Mack A & Robitzki A (2000) The key role of

butyr-ylcholinesterase during neurogenesis and neural

disor-ders: an antisense-5¢butyrylcholinesterase-DNA study

Prog Neurobiol 60, 607–628

17 Paraoanu LE, Steinert G, Koehler A, Wessler I &

Layer PG (2007) Expression and possible functions of

the cholinergic system in a murine embryonic stem cell

line Life Sci 80, 2375–2379

18 Johnson G, Swart C & Moore SW (2008)

Non-enzy-matic developmental functions of acetylcholinesterase –

the question of redundancy FEBS J 275, 5129–5138

19 Greenfield S, Zimmermann M & Bond CE (2008)

Non-hydrolytic functions of acetylcholinesterase The

significance of C-terminal peptide FEBS J 275, 604–

611

20 Silveyra MX, Evin G, Montenegro MF, Vidal CJ,

Martı´nez S, Culvenor JG & Sa´ez-Valero J (2008)

Prese-nilin-1 interacts with acetylcholinesterase and alters its

enzymatic activity and glycosylation Mol Cell Biol 28,

2908–2919

21 Mor I, Bruck T, Greenberg D, Berson A, Schreiber L,

Grisaru D & Soreq H (2008) Alternate AChE-R

vari-ants facilitate cellular metabolic activity and resistance

to genotoxic stress through enolase and RACK1

inter-actions Chem Biol Interact 175, 11–21

22 Perry C, Pick M, Podoly E, Gilboa-Geffen A, Zimmerman G, Sklan EH, Ben-Shaul Y, Diamant S & Soreq H (2007) Acetylcholinesterase⁄ C terminal binding protein interactions modify Ikaros functions, causing

T lymphopenia Leukemia 21, 1472–1480

23 Gong X, Ye W, Zhou H, Ren X, Zhigang L, Zhou W,

Wu J, Gong Y, Ouyang Q, Zhao X et al (2009) Ran-BPM is an acetylcholinesterase-interacting protein that translocates into the nucleus during apoptosis Acta Biochim Biophys Sin 41, 883–891

24 Darreh-Shori T, Modiri N, Blennow K, Baza S, Kamil C, Ahmed H, Andreasen N & Nordberg A (2010) The apolipoprotein E varepsilon4 allele plays pathological roles in AD through high protein expres-sion and interaction with butyrylcholinesterase Neuro-biol Agingdoi:10.1016/j.neurobiolaging.2009.07.015

25 Darreh-Shori T, Forsberg A, Modiri N, Andreasen

N, Blennow K, Kamil C, Ahmed H, Almkvist O, Lanstro¨ng B & Nordberg A (2010) Differential levels

of apolipoprotein E and butyrylcholinesterase show strong association with pathological signs of Alzheimer’s disease in the brain in vivo

Neurobiol Aging doi:10.1016/j.neurobiolaging.2010.04 028

26 Vidal CJ (2005) Expression of cholinesterases in brain and non-brain tumours Chem Biol Interact 157–158, 227–232

27 Perry C, Sklan EH, Birikh K, Shapira M, Trejo L, Eldor A & Soreq H (2002) Complex regulation of acetylcholinesterase gene expression in human brain tumors Oncogene 21, 8428–8441

28 Calaf GM, Parra E & Garrido F (2007) Cell prolifera-tion and tumor formaprolifera-tion induced by eserine, an acetyl-cholinesterase inhibitor, in rat mammary gland Oncol Rep 17, 25–33

29 Ruiz-Espejo F, Cabezas-Herrera J, Illana J, Campoy FJ, Mun˜oz-Delgado E & Vidal CJ (2003) Breast cancer metastasis alters acetylcholinesterase activity and the composition of enzyme forms in axillary lymph nodes Breast Cancer Res Treat 80, 105–114

30 Montenegro MF, Ruiz-Espejo F, Campoy FJ, Mun˜oz-Delgado E, Pa´ez de la Cadena M, Rodrı´guez-Berrocal

FJ & Vidal CJ (2006) Cholinesterases are down-expressed in human colorectal carcinoma Cell Mol Life Sci 63, 2175–2182

31 Martı´nez-Moreno P, Nieto-Cero´n S, Torres-Lanzas J, Ruiz-Espejo F, Tovar-Zapata I, Martı´nez-Herna´ndez P, Rodrı´guez-Lo´pez JN, Vidal CJ & Cabezas-Herrera J (2006) Cholinesterase activity of human lung tumours varies according to their histological classification Carcinogenesis 27, 429–436

32 Suda A (1986) A histochemical localization of acetyl-cholinesterase and acetyl-cholinesterase activities in mamma-lian kidneys Acta Histochem 79, 107–114