Immunology and Cell Biology

(2003)

81

, 275–282

Research Article

mRNA of MUC2 is stimulated by IL-4, IL-13 or TNF-

α

through a mitogen-activated protein kinase pathway in human colon cancer cells

J U N I W A S H I T A ,

1

Y U K I T A S A T O ,

1

H I R O K O S U G A Y A ,

2

N A G A T O M O T A K A H A S H I ,

2

H I R O S H I S A S A K I

2

a n d T A T S U Y A A B E

1

1

Molecular Biology, Akita Prefectural University,

2

Department of Parasitology, Akita University School of Medicine, Akita, Japan

Summary

MUC2 mucin is a secretory glycoprotein which is produced from the intestinal goblet cells and is amajor component of the intestinal epithelial mucus. The biological function of MUC2 mucin is considered to be theprotection of intestinal epithelial surface, whereas the regulatory mechanism of MUC2 mucin production in immuneresponse is not completely understood. We have studied the effects of cytokines, IL-4, IL-13 and TNF-

α

, on theregulation of MUC2 mRNA in the human colonic cancer cell lines, LS174T and HT29. The quantitative reversetranscription-polymerase chain reaction showed that single addition of IL-4, IL-13 and TNF-

α

to cell cultureinduced about two-fold increase of MUC2 mRNA level in LS174T cells. Interleukin-4 and IL-13 activatedphosphorylation of mitogen-activated protein kinase in LS174T cells. A specific inhibitor of mitogen-activatedprotein kinase pathway, U0126, totally inhibited the increase of MUC2 mRNA by IL-4 or IL-13 in those cells.Therefore, mitogen-activated protein activation of kinase is required for the increase of MUC2 mRNA by IL-4 orIL-13 in LS174T cells. In contrast to LS174T cells, only TNF-

α

increased MUC2 mRNA through a mitogen-activated protein kinase pathway in HT29 cells that express low levels of MUC2 mRNA. These findings sustain anovel phenomenon that MUC2 mRNA expression is differently controlled by IL-4, IL-13, or TNF-

α

in LS174T andHT29 cells, whereas the mitogen-activated protein kinase pathway plays a role in the MUC2 mRNA expressioninduced by those cytokines in both cell lines.

Key words

:

cytokine, HT29 cell, LS174T cell, mitogen-activated protein kinase, MUC2.

Introduction

Vertebrates possess a single tube formed by the mouth,oesophagus, stomach and intestine. The intestine has appar-ently contrasting roles, which are absorption of nutrientsthrough the epithelia and rejection of harmful pathogens thatattempt to invade through the epithelia. The surface of theintestinal epithelia is covered with a viscous mucus layer ofwhich mucin is the major constituent.

1,2

Intestinal mucin,which is mainly produced and secreted by intestinal gobletcells, is a large glycoprotein composed of a core proteincovered with a large amount of O-linked glycosides.

3,4

Currently, 17 subtypes of human mucin core proteins havebeen isolated from different organs, and are collectivelyreferred to as the MUC family.

4–8

The MUC family can beclassified into two groups based on their characteristic homo-logy. One of them is the secreted and mucus-forming mucins,MUC2, MUC5B, MUC5AC and MUC6, which are encodedat the chromosome locus 11p15. Another group is themembrane-bound mucins which are encoded at the chromo-some loci 7q22, 3q or 1q21.

8

Different subtypes of the MUCcore proteins are specifically expressed in different humanorgans and cells. MUC5AC mucin is predominant in the

airway surface, whereas MUC2 and MUC3 mucins are highlyexpressed in the small intestine.

5,9,10

MUC2 is a major mucinof the human colon, and is produced and secreted fromcolonic goblet cells but not from colonic epithelial cells.

11

Carcinogenesis and some types of pathogens are known tomodulate the production and secretion of MUC2 mucin. Inthe majority of colonic carcinomas, MUC2 mRNA expressionis reduced to low or undetectable levels,

10,12

whereas a highlevel of MUC2 expression is detected in the so-called muci-nous carcinomas.

13

Such differential expression of MUC2mucin in the various types of colonic carcinoma argues for asignificant relationship between colonic carcinogenesis andMUC2 expression. Recently, the MUC2 gene targeted micedemonstrated a direct evidence that MUC2 mucin plays acrucial role in the induction of colorectal cancer.

14

Physiological secretion of intestinal mucin includingMUC2 is generally considered to be controlled by the para-sympathetic nervous system. It is known that gene expressionof the secreted mucins encoded at chromosome locus 11p15is regulated by several cytokines such as IL-1

β

, IL-6, TNF-

α

,IL-4, IL-9 or IL-13.

15

However, the mechanisms regulatingintestinal mucin production in immune response is still notcompletely understood. Interleukin-4 has been shown toinduce MUC2 gene expression in cultured airway epithelialcells, and MUC5 gene expression in mice.

16

In addition tothese reports on the effects of cytokines several recent reportshave described the signal transduction pathway of mucinproduction. Expression of colonic

MUC2

gene has been

Correspondence: Dr Tatsuya Abe, Molecular Biology, AkitaPrefectural University, Akita 010–0195, Japan. Email: abetats@akita-pu.ac.jp

Received 29 November 2002; accepted 14 March 2003.

276

J Iwashita

et al

.

demonstrated to be directly regulated by methylation of itspromoter domain

17

and the transcription factor Sp1.

18

Aspecific inhibitor of mitogen-activated protein (MAP) kinasekinase, U0126, has been found to reduce expression ofsialomucin complex, MUC4, in rat adenocarcinoma cells.

19

Furthermore, synthesis of MUC5AC mucin is stimulated byoxidative stress through transactivation of epidermal growthfactor receptor and subsequent activation of MAP kinase inthe human pulmonary carcinoma cell line NCI-H292.

20

Here, we have focused on the mechanisms regulatingmucin production in immune response, and have examinedthe effects of cytokines on MUC2 mRNA expression in thetwo different human colonic cancer cell lines, LS174T, andHT29 (HTB-38). LS174T is a goblet cell-like colon-cancercell line and produces high-levels of MUC2 mucin continuously.On the other hand, the epithelial-like colon-adenocarcinomacell line, HT29, produces low-levels of MUC2 mucin. Wereport that MUC2 mRNA expression in both colonic cell linesis stimulated by IL-4, IL-13 or TNF-

α

through a MAP kinasepathway.

Materials and methods

Cell lines and reagents

The human colon cancer cell lines, LS174T and HT29 (HTB-38),were obtained from Dainippon Pharmaceutical Co. (Osaka, Japan).Culture mediums, RPMI-1640, MEM and McCoy5A, were obtainedfrom Sigma Aldrich Co. (Tokyo, Japan), and fetal bovine serum(FBS) was from Cansera International, Canada. LS174T cells werecultured in 10% FBS-MEM-1% nonessential amino acids-100 unit/mL penicillin-100

µ

g/mL streptomycin (All reagents from GibcoOriental, Japan) in a 5% CO

2

incubator. HT29 cells were cultured in10% FBS-McCoy5A-100 unit/mL penicillin-100

µ

g/mL strepto-mycin. LS174T and HT29 cells were collected before confluence,following 6–7 days of subculture, and were subsequently used forexperiments. Phorbol myristate acetate (PMA) was obtained fromSigma (Tokyo, Japan), and the MAP kinase kinase specific inhibitor,U0126, was obtained from Merck (Tokyo, Japan). Recombinanthuman IL-13 was obtained from Serotec Inc (NC, USA), recombinanthuman IL-4 was from R & D Systems (MI, USA), and recombinanthuman TNF-

α

was from PharMingen (San Diego, USA).

Exposure of cells to cytokines

LS174T or HT29 cells (2

×

10

6

cells in 2 mL) were incubated withdifferent concentrations of cytokines (IL-13, IL-4, TNF-

α

) in serum-free medium in 24-well plates. The cells were cultured in a 5% CO

2

incubator for indicated time. Phorbol myristate acetate (20 ng/mL)was added in culture medium for 15 min for activation of MAPkinase. U0126 (30

µ

mol/L) was used as MAP kinase kinase inhibitor.

Assay of mucin

LS174T cells (10

6

cells in one mL) were cultured with IL-13 (0.1 and1 ng/mL) for 2, 3 or 6 days in a 24-well plate in triplicate. Cells wereharvested in a tube and dispersed by syringing with 25G needle. Aftercentrifugation, the supernatant was collected. Mucin was assayed bya slot-blot method using BIO-DOT SF apparatus (Bio-Rad, USA).Samples of mucin were blotted onto an Immobilon (Millipore, USA)membrane by aspiration. The blotted membrane was incubated with5% FBS-Tris buffered saline, and then incubated with a dilutedperoxidase-labelled soybean lectin solution (1 : 1000, Wako, Japan).

After washing twice with 0.05% Tween 20-tris-buffered saline (TBS)for 5 min and once with TBS, the membrane was incubated with acolour developer solution (4-chloro-1-naphtol, Bio-Rad) to detect theenzymatic reaction. The colour bands on the membrane were detectedwith a scanner (Seiko Epson, Japan). The amount of sample mucinwas analysed using NIH Image with a standard curve obtained froma known amount of mucin.

Isolation of ribonucleic acid

Cellular RNA was isolated with a TRIzol Reagent kit (Invitrogen,Japan). Cells (10

4

), collected by centrifugation, were mixed with0.8 mL of the TRIzol Reagent, homogenized by vigorous pipetting,and left at room temperature for 5 min. After addition of 0.06 mL ofchloroform, cells were incubated at room temperature for 3 min.Following centrifugation at 12 000

×

g for 10 min, the aqueous phasewas collected, mixed with 0.4 mL of isopropyl alcohol, and left atroom temperature for 10 min. After centrifugation, the precipitatedRNA was collected, washed with 70% ethanol, and dissolved inRNase-free water. The RNA was treated with DNase I (Nippon Gene,Japan) for 30 min and extracted by standard methods. The DNA-freeRNA was used for the detection of mRNAs by real-time reversetranscription-polymerase chain reaction (RT-PCR).

Reverse transcription-polymerase chain reaction

MUC2 mRNA, as well as some cytokine receptor mRNAs, wasassayed by RT-PCR using a real-time PCR LightCycler System V3(Roche Molecular Biochemicals, Germany) and a QuantiTect SYBRGreen RT-PCR Kit (Qiagen, Japan). Total RNA (30 ng), forwardprimer and reverse primer (10 pmol each) were mixed with the kitsolution containing reverse transcriptase, deoxyribonucleoside tri-phosphates (dNTPs), Taq DNA polymerase and fluorescence dyes ina 20

µ

L solution. After one cycle of reverse transcription at 50

°

C for20 min and activation of Taq polymerase, PCR was performed for 55cycles. The PCR was a cycle of 94

°

C for 15 s, 64

°

C for 15 s and 72

°

Cfor 20 s. All reactions were carried out in triplicate. The amounts ofMUC2 and receptor mRNAs were obtained from a standard curve,and corrected by amounts of

β

-actin mRNA. To visualize the relativeamount of PCR products, PCR was stopped at the log-linear phaseafter 32 cycles for MUC2 and 38 cycles for

β

-actin in someexperiments. The PCR products were detected by electrophoresis ona 1.5% agarose gel-containing ethidium bromide. The PCR primersused are as follows:hMUC2 forward: 5

′

-CAGCACCGATTGCTGAGTTG-3

′

,hMUC2 reverse: 5

′

-GCTGGTCATCTCAATGGCAG-3

′

;h

β

-actin forward: 5

′

-CCAACCGCGAGAAGATGAC-3

′

,h

β

-actin reverse: 5

′

-GGAAGGAAGGCTGGAAGAGT-3

′

;hIL-4 receptor

α

forward: 5

′

-CAAGCTCTTGCCCTGTTTTC-3

′

,hIL-4 receptor

α

reverse: 5

′

-TCCTCCTCCTCACACTCCAC-3

′

;hIL-13 receptor

α

forward: 5

′

-TCCCAGTGTAGCACCAATGA-3

′

,hIL-13 receptor

α

reverse: 5

′

-CCAGGCTTCTGTGCCAATAG-3

′

;hTNF-

α

receptor 120a forward: 5

′

-ACCGGCATTATTGGAGT-GAA-3

′

,hTNF-

α

receptor 120a reverse: 5

′

-TCTGGGGTAGGCACAACTTC-3

′

,hTNF-

α

receptor 120b forward: 5

′

-CTCCAACACGACTTCATCCA-3

′

and hTNF-

α

receptor 120b reverse: 5

′

-CAGTGCTGGGTTCT-GGAGTT-3

′

.

Sequence analysis of PCR products

Reverse transcription-polymerase chain reaction products werecollected by ethanol precipitation, and their nucleotide sequencesdetermined using an ABI 310 DNA sequencer (Perkin Elmer, USA)

MUC2 mRNA through MAP kinase pathway

277

and a BigDye Terminator Cycle Sequencing Ready Reaction Kit(Applied Biosystems, Japan). The reaction mixtures (10

µ

L), each ofwhich contained 10 ng of PCR product, 1.6 pmol of forward orreverse primer and 4

µ

L of the kit reagent, were subjected to 25 PCRsequencing cycles. The PCR was a cycle of 96

°

C for 30 s, 50

°

C for15 s and 60

°

C for 4 min.

Antibodies

Phospho-STAT6 (Tyr641) polyclonal antibody and phospho-p44/p42MAP kinase (Thr202/Tyr204) E10 monoclonal antibody wasobtained from Cell Signalling Technology (MA, USA). Phospho-STAT6 (Tyr641) antibody detects STAT6 only when activated withphosphorylation at Tyr641. Phospho-p44/p42 MAP kinase (Thr202/Tyr204) E10 antibody detects p44 and p42 MAP kinases (ERK1 andERK2) only when activated by dually phosphorylated at Thr202 andTyr204.

Western blot analysis

Treated cells were collected, dissolved in Laemmli’s sample buffer

21

and heated at 98

°

C for 3 min. Aliquots were then subjected to 7.5%and 10% SDS-polyacrylamide gel electrophoresis for detection ofSTAT6 and MAP kinase, respectively. After electrophoresis, proteinson the gel were transferred to nitrocellulose membrane (HybondECL: Amersham Pharmacia Biotech, USA) by electrophoresis at100 mA for 45 min using a transfer apparatus (Cima Biotech, USA)with a transfer buffer composed of 20 mmol/L Tris, 150 mmol/Lglycine and 20% methanol. Following transfer, the membrane wasincubated in a blocking buffer of TBS-T (20 mmol/L Tris-HCl,150 mmol/L NaCl, 0.1% Tween-20, pH 7.5) containing 4% skimmedmilk (Gibco Oriental, Japan) at room temperature for 1 h. Themembrane was incubated with antiphospho-STAT6 (Tyr641) anti-body (1 : 5000) or antiphospho-p44/p42 MAP kinase (Thr202/Tyr204) E10 monoclonal antibody (1 : 5000) in blocking buffer for1 h at room temperature. The antiphospho-MAP kinase E10 mAbreacts with Thr202/Tyr204-phosphorylated p44 MAP kinase and p42MAP kinase, but does not cross-react with non-phosphorylated MAPkinase or with a single residue-phosphorylated MAP kinase. Afterwashing with TBS-T, the membrane was then incubated for 1 h withgoat antirabbit IgG (H + L) antibody conjugated with horseradishperoxidase (1 : 5000; Promega, USA) for STAT6 detection, or goatantimouse IgG antibody conjugated with horseradish peroxidase(1 : 5000; Promega, USA) for MAP kinase detection. After mem-brane washing with TBS-T, the enzyme reactions were detected by anenhanced chemiluminescence assay using an ECL kit (AmershamCo., USA) and a LAS-1000 plus image analyser (Fujifilm, Japan).

Results

IL-13 stimulates mucin protein secretion from LS174T human colon cancer cells

Interleukin-13 is required to develop intestinal goblet cellhyperplasia that is induced by intestinal parasite infection.

15

We examined the effects of IL-13 on mucin secretion inLS174T cells, a goblet cell-like colon-cancer cell line, whichproduce mucin continuously. LS174T cells were culturedwith IL-13 for 2, 3 or 6 days. The number of cells increasedgradually with culture period, but no difference in cellnumber was detected between the IL-13 treated and untreatedcontrol cells. However, the amount of mucin secreted into theculture medium was increased by addition of 1 ng/mL IL-13

(Fig. 1). Similar experiments with IL-4 and TNF-

α

showedthat they also stimulated mucin secretion from LS174T cells(data not shown).

IL-4, IL-13 and TNF-

α

stimulate MUC2 mRNA expression in LS174T cells

The major subtype of mucin produced in LS174T cells isMUC2 mucin. To examine the effects of IL-4, IL-13 andTNF-

α

on MUC2 mRNA levels in LS174T cells, we per-formed the real-time RT-PCR. The levels of MUC2 mRNA intotal RNA were adjusted with

β

-actin mRNA as a control.The identities of the amplified MUC2 and

β

-actin PCRproducts were confirmed by DNA sequence analysis. Thelevels of

β

-actin mRNA were unaffected by IL-4, IL-13, orTNF-

α

, whereas MUC2 mRNA expression in LS174T cellswas peaked and two-fold increased following 24 h treatmentwith these cytokines (Fig. 2). We examined the IL-4, IL-13and TNF-

α

receptor mRNA levels in the LS174T cells todetermine whether this increase in MUC2 mRNA expressionwas due to an increase in the number of cytokine receptors ofthe cells. Although mRNAs of these cytokine receptors weredetectable by RT-PCR, no increases in their levels weredetected in response to IL-4, IL-13, or TNF-

α

treatment.

IL-4 and IL-13 activate phosphorylation of STAT6 and MAP kinase in LS174T cells

A major signal transduction pathway activated by IL-4 orIL-13 in human cells is the JAK/STAT pathway. The activa-tion of JAK/STAT pathway by IL-4 or IL-13 is representedby phosphorylation of STAT6. We therefore, examined thephosphorylation of STAT6 in LS174T cells after cytokinestimulation. The LS174T cells were incubated with IL-4,IL-13, or TNF-

α

for 24 h, followed by detection of phos-phorylated STAT6 by Western blot analysis using the anti-phospho STAT6 antibody. Interleukin-4 and IL-13 activatedthe phosphorylation of STAT6 within 10–30 min, whereasTNF-

α

did not affect the phosphorylation of STAT6 inLS174T cells (Fig. 3a).

Figure 1

Effect of IL-13 on mucin secretion in LS174T cells.LS174T cells were cultured with IL-13 (0.1 and 1 ng/mL) for 2, 3or 6 days in triplicate. After centrifugation, mucin in the super-natant was measured. Sample without IL-13, (

�

); sample incubatedwith 0.1, ( ); and sample incubated with 1 ng/mL of IL-13, (

�

).The mean values (

±

SD) for each sample are shown.

278

J Iwashita

et al

.

We subsequently examined the phosphorylation of MAPkinase, a key molecule in another major signal transductionpathway. Western blot analysis, using an antiphospho MAPkinase antibody, demonstrated that MAP kinase was phos-phorylated within 5–30 min by IL-4 and IL-13, but wasunaffected by TNF-

α

(Fig. 3b).

MUC2 mRNA expression is stimulated through a MAP kinase pathway in LS174T cells

The novel finding that the MAP kinase pathway could beactivated by IL-4 and IL-13 in LS174T cells prompted us toexamine the correlation between activation of MAP kinaseand induction of MUC2 mRNA expression by thesecytokines. Mitogen-activated protein kinase is phosphor-ylated and activated by a MAP kinase kinase that can bespecifically inhibited by an inhibitor, U0126, which thereforeinhibits the MAP kinase pathway. Incubation of LS174T cellswith IL-4 and IL13, together with U0126 inhibitor, demon-strated that the stimulation of MUC2 mRNA expression byIL-4 and IL13 in LS174T cells could be totally inhibited bythe inhibitor, U0126 (Fig. 4). Therefore, IL-4 and IL-13induce MUC2 mRNA expression through the MAP kinasepathway in LS174T cells.

Tumour necrosis factor-

α

stimulates MUC2 mRNA expression through MAP kinase phosphorylation in HT29 cells

We examined the MUC2 mRNA stimulation induced bycytokines in another colonic cancer cell line, HT29 (HTB-38), to see if MAP kinase pathway is generally responsible forthe stimulation. LS174T cells produce high-levels of MUC2mucin continuously, whereas HT29 cells produce a low-levelof MUC2 mucin. HT29 cells were cultured with IL-4, IL-13,or TNF-

α

for 16 and 72 h to measure MUC2 mRNA levels. Incontrast to LS174T cells, IL-4 and IL-13 did not stimulateMUC2 mRNA levels in HT-29 cells. Only TNF-

α

of 16 hculture induced two-fold increase of MUC2 mRNA expres-sion (Fig. 5). Although apparent accumulation of MUC2mRNA was not detected by incubation with IL-4 or IL-13 inHT29 cells, these cytokines induced phosphorylation ofSTAT6 and MAP kinase. A different point in HT29 cellsfrom LS174T cells was that the activation of MAP kinase

Figure 2

Reverse transcription-polymerase chain reaction (RT-PCR) of MUC2 mRNA and the effect of cytokines on MUC2mRNA expression in LS174T cells. LS174T cells were culturedwith 1 ng/mL of IL-13 (�), IL-4 ( ) or TNF-α ( ) for 24 h.Ribonucleic acid was extracted at 24 h for measurement of MUC2mRNA levels in triplicate. The open bar represents control, whichis incubated without cytokines (�). Data represent the foldincrease against the MUC2 level of no stimulation. The PCRproducts of log-linear phase are visualized below.

Figure 3 Activation of STAT6 or mitigen-activated protein(MAP) kinase by cytokines in LS174T cells. (a) Activation ofSTAT6 in LS174T cells was detected with an antiphospho STAT6antibody. LS174T cells were incubated with 1 ng/mL of IL-4,IL-13 or TNF-α for the time indicated. Activated STAT6 isindicated by the arrows. (b) Activation of MAP kinase wasdetected with an antiphospho MAP kinase antibody. Upper andlower bands of MAP kinase are phosphorylated p44 MAP kinase(ERK1) and phosphorylated p42 MAP kinase (ERK2), respec-tively. Phosphorylated MAP kinase was indicated by the arrows.A positive control (P) treated with the MAP kinase activator,phorbol myristate acetate (PMA) (20 ng/mL), for 15 min is shown.

MUC2 mRNA through MAP kinase pathway 279

phosphorylation was induced within 5–25 min by incubationwith TNF-α (Fig. 6). The MUC2 mRNA expression in HT29cells was increased by TNF-α, therefore, we examined thecorrelation of MAP kinase activation and MUC2 mRNAexpression in HT29 cells. An inhibitor of MAP kinase path-way, U0126, showed that the stimulation of MUC2 mRNAlevel by TNF-α could be inhibited by addition of U0126 inHT29 cell culture (Fig. 7). Therefore, TNF-α induces MUC2mRNA expression through the MAP kinase pathway in HT29cells.

These results sustain a novel phenomenon that MUC2mRNA expression is differently controlled by IL-4, IL-13, orTNF-α in LS174T and HT29 cells, whereas MAP kinasepathway plays a role in the MUC2 mRNA expression inducedby those cytokines in both cell lines.

Discussion

The production and secretion of mucin from intestinal gobletcells are known to be up-regulated in particular diseases orparasitic infections.2,22 The up-regulation of mucin is probablycontrolled by immune response, but cytokine signals andintracellular signal transduction for mucin regulation are notcompletely clear. We examined the effects of IL-4, IL-13 andTNF-α on the expression of MUC2 mRNA in the humancolon cancer cell lines, LS174T and HT29. We showed thatIL-4 and IL-13 stimulated MUC2 protein secretion and

MUC2 mRNA expression in LS174T cells. This stimulationof MUC2 mRNA expression was inhibited to a non-stimulated level by a MAP kinase kinase-specific inhibitor,U0126, which inhibits the phosphorylation of MAP kinase.The activation of MAP kinase pathway is therefore required forthe stimulation of MUC2 mRNA expression by IL-4 andIL-13 in LS174T cells. The JAK/STAT pathways may beused in part for the stimulation of MUC2 mucin expression byIL-4 and IL-13 in LS174T cells, because these cytokinesactivated the phosphorylation of STAT6 in those cells. But,our results show that the MAP kinase pathway is moreimportant than the JAK/STAT pathway for the stimulation ofMUC2 mRNA expression by IL-4 and IL-13 in LS174T cells.The IL-4 and IL-13-induced gene expression through a MAPkinase pathway is unusual. Interleukin-4 and IL-13 usuallyfunction through the JAK/STAT signal transduction path-way.23 As for stimulation of MAP kinase pathway by IL-4,Wery-Zennaro et al.24 reported that IL-4 stimulates IL-6 pro-duction through the p38 MAP kinase-dependent pathway in akeratinocyte cell line. It has been reported that the directinteraction of epithelial cell lines and the blue pus bacillus,Pseudomonas aeruginosa, stimulates MUC2 production fromthe cells through a Src-dependent Ras-MAP kinase-pp90rskpathway.25 As far as we are aware, however, there is no reportthat IL-4 or IL-13 stimulates MUC2 mRNA expressionthrough a MAP kinase pathway.

We found that TNF-α also induced an increase of MUC2mRNA expression in LS174T cells without activation of

Figure 4 Effect of a mitigen-activated protein (MAP) kinasekinase inhibitor on MUC2 mRNA expression in LS174T cells.LS174T cells were incubated with IL-4 or IL-13 (1 ng/mL) withor without the MAP kinase kinase inhibitor, U0126 (30 µmol/L),for 24 h ( ) or 72 h (�). Ribonucleic acid was extracted formeasurements of MUC2 mRNA in triplicate. Data represent thefold increase against the MUC2 level of no stimulation for 24 h.The log-linear phase PCR-products of 24 h incubation are visual-ized below.

Figure 5 Effects of cytokines on MUC2 mRNA expression inHT29 cells. HT29 cells were cultured with 5 ng/mL of IL-4, IL-13or TNF-α for 16 h (left bar) or 72 h (right grey bar). MUC2mRNA was measured by reverse transcription-polymerase chainreaction (RT-PCR) in triplicate. Data represent the fold increaseagainst the MUC2 RNA level of no stimulation for 16 h. The log-linear phase PCR-products of 16 h culture are visualized below.

280 J Iwashita et al.

STAT6 or MAP kinase. Up-regulation of MUC5AC mucin byTNF-α has been reported in HT29-MTX cells.26 The TNF-α-induced reactive oxygen species stimulate MUC5AC synthe-sis through transactivation of epidermal growth factor recep-tors and subsequent activation of MAP kinase pathway.20 Ourresults, however, show that TNF-α can induce the up-regulation of MUC2 mRNA through a pathway other thanMAP kinase pathway in LS174T cells.

We used another colonic cancer cell line, HT29 (HTB-38),to see if a MAP kinase pathway is generally responsible forthe stimulation of MUC2 mRNA expression. In contrast toLS174T cells, neither IL-4 nor IL-13 stimulated MUC2mRNA expression in HT29 cells. Only TNF-α stimulatedMUC2 mRNA expression through a MAP kinase pathway inHT29 cells. Tumour necrosis factor-α is known to activatecomplex signal transduction pathways that induce contraryeffects in different cells.27 The differential effects of TNF-α,in which a MAP kinase pathway was activated for MUC2mRNA expression in HT29 cells but not in LS174T cells,

may be attributable to the different characteristics of thesignal transduction pathways of these cells. The cell linesLS174T and HT29 are obviously different in theirmorphology28 and in their feature of mucin production. TheLS174T cell lines are goblet cell-like colonic cancer cells,which produce a large amount of mucin including MUC2spontaneously. The HT29 cell lines are colonic adenocarci-noma cells which express a low level of MUC2 mRNA.28–30

Despite such differences between two cell lines, our experi-ments using the MAP kinase kinase inhibitor, U0126, clearlyshowed that stimulation of the MUC2 mRNA expression byeither IL-4, IL-13 or TNF-α is mediated through a MAPkinase pathway in both cell lines.

Recently, the importance of MUC2 mucin in vivo has beenreported in mice. The gene targeted MUC2 deficient micefrequently develop adenomas in the small intestine thatprogress to invasive adenocarcinoma, as well as rectaltumours.14 Modulation of MUC2 mucin production is wellknown in some types of cancer.10,12,13 Activation of the MAPkinase pathway is indispensable for cell proliferation orcarcinogenesis. The up-regulation of MUC2 mucin produc-tion by cytokines through MAP kinase activation may be aninteresting point to be studied for the mechanism of carcino-genesis.

Figure 6 Activation of STAT6 or mitogen-activated protein(MAP) kinase with cytokines in HT29 cells. (a) Activation ofSTAT6 was detected with antiphospho STAT6 antibody in HT29cells that were incubated with 5 ng/mL of I IL-4, L-13 or TNF-αfor the time indicated. Activated STAT6 was indicated by thearrows. (b) Activation of MAP kinase was detected with anantiphospho MAP kinase antibody. Phosphorylated MAP kinasewas indicated by the arrows. Upper and lower bands of MAPkinase are phosphorylated p44 MAP kinase (ERK1) and phos-phorylated p42 MAP kinase (ERK2), respectively. A positivecontrol (P) treated with a MAP kinase activator, phorbol myristateacetate (PMA) (20 ng/mL), for 15 min is shown.

Figure 7 Effect of mitogen-activated protein (MAP) kinasekinase inhibitor on MUC2 mRNA expression in HT29 cells. HT-29cells were incubated with 1 ng/mL or 5 ng/mL of TNF-α with orwithout the MAP kinase kinase inhibitor, U0126 (30 µmol/L), for16 h. Ribonucleic acid was extracted for reverse transcription-polymerase chain reaction (RT-PCR) measurements of MUC2mRNA in triplicate. The PCR products of log-linear phase arevisualized below.

MUC2 mRNA through MAP kinase pathway 281

Immunological regulation of mucin production hasattracted considerable attention in the study of the airwaysince IL-13 was demonstrated to be required for allergicasthma31 and to be critical to the stimulation of mucusproduction in allergic asthma.32 The level of gene expressionof MUC5AC, which is the predominant subtype in theairway,33 is increased by IL-4 in the human pulmonarycarcinoma cell line, NCI-H292.16 In this same cell line,MUC5AC gene expression is also found to be increased byTNF-α through a MAP kinase pathway.18,20 We concludefrom these results, and those from our present studies that theexpression of some MUC subtypes is regulated by cytokinesthrough a MAP kinase pathway.

Acknowledgements

This work was supported in part by a Grant-in-Aid forScientific Research (C) from JSPS (No. 12670236). We thankDr Shohab Youssefian, Akita Prefectural University, for hiscritical review of this manuscript.

References

1 Szentkuti L, Riedesel H, Enss ML, Gaertner K, Engelhardt WV.Pre-epithelial mucus layer in the colon of conventional andgerm-free rats. Histochem. J. 1990; 22: 491–7.

2 Mack DR, Michail S, McDougal SWL, Hollingsworth MA. Pro-biotics inhibit enteropathogenic E. coli adherence in vitro byinducing intestinal mucin gene expression. Am. J. Physiol. 1999;276: G941–50.

3 Deplanck B, Gaskins HR. Microbial modulation of innatedefense: goblet cells and the intestinal mucus layer. Am. J. Clin.Nutr. 2001; 73: 1131S–41S.

4 Gendler SJ, Spicer AP. Epithelial mucin genes. Annu. Rev.Physio. 1995; 57: 607–34.

5 Meezaman D, Charles P, Daskal E, Polymeropoulos MH,Martin BM, Rose MC. Cloning and analysis of cDNA encodinga major airway glycoprotein, human tracheobronchial mucin(MUC5). J. Biol. Chem. 1994; 269: 12 932–9.

6 Williams SJ, McGuckin MA, Gotley DC, Eyre HJ,Sutherland GR, Antalis TM. Two novel mucin genes down-regulated in colorectal cancer identified by differential display.Cancer. Res. 1999; 59: 4083–9.

7 Williams SJ, Wreschner DH, Tran M, Eyre HJ, Sutherland GR,McGuckin MA. Muc13, a novel human cell surface mucinexpressed by epithelial and hemopoietic cells. J. Biol. Chem.2001; 276: 18327–36.

8 Dekker J, Rossen JWA, Rossen HA, Einerhand AWC. The MUCfamiy: an obituary. TRENDS Biochem. Sci. 2002; 27: 126–31.

9 Chang SK, Dohrman AF, Basbaum CB et al. Localization ofmucin (MUC2 and MUC3) messenger RNA and peptide expres-sion in human normal intestine and colon cancer. Gastroenterol-ogy 1994; 107: 28–36.

10 Weiss AA, Babyatsky MW, Ogata S, Chen A, Itzkowitz SH.Expression of MUC2 and MUC3 mRNA in human normal,malignant, and inflammatory intestinal tissues. J. Histochem.Cytochem. 1996; 44: 1161–6.

11 Winterford CM, Walsh MD, Leggett BA, Jass JR. Ultrastructurallocalization of epithelial mucin core proteins in colorectal tis-sues. J. Histochem. Cytochem. 1999; 47: 1063–74.

12 Sylvester PA, Myerscough N, Warren BF et al. Differentialexpression of the chromosome 11 mucin genes in colorectalcancer. J. Pathol. 2001; 195: 327–35.

13 O’Connell JT, Shao ZM, Drori E, Basbaum CB, Barsky SH.Altered mucin expression is a field change that accompaniesmucinous (colloid) breast carcinoma histogenesis. Hum. Pathol.1998; 29: 1517–23.

14 Velcich A, Yang W, Hayer J et al. Colorectal cancer in micegenetically deficient in the mucin Muc2. Science 2002; 295:1726–9.

15 Van Seuningen I, Pigny P, Perrais M, Porchet N, Aubert JP.Transcriptional regulation of the 11p15 mucin genes: Towardsnew biological tools in human therapy, in inflammatory diseasesand cancer? Front. Biosci. 2001; 6: D1216–34.

16 Dabbagh K, Takeyama K, Lee HM, Ueki IF, Lausier JA,Nadel JA. IL-4 induces mucin gene expression and goblet cellmetaplasia in vitro and in vivo. J. Immunol. 1992; 162: 6233–7.

17 Gratchev A, Siedow A, Bumke-Vogt C et al. Regulation of theintestinal mucin MUC2 gene expression in vivo: evidence for therole of promoter methylation. Cancer Lett. 2001; 168: 71–80.

18 Perrais M, Pigny P, Copin M-C, Aubert J-P, Seuningen IV.Induction of MUC2 and MUC5AC mucin by factors of theepidermal growth factor family is mediated by EGF-R./Ras/Raf/MAPK signaling cascade and Sp1. J. Biol. Chem. 2002; 277:32258–67.

19 Zhu X, Price-Schiavi SA, Carraway KL. Extracellular regulatedkinase (ERK) -dependent regulation of sialomucin complex/ratMuc4 in mammary epithelial cells. Oncogene 2000; 19:4354–61.

20 Takeyama K, Dabbagh K, Jeong-Shim J, Dao-Pick T, Ueki IF,Nadel JA. Oxidative stress causes mucin synthesis via transacti-vation of epidermal growth factor receptor: role of neutrophils.J. Immunol. 2000; 164: 1546–52.

21 Laemmli UK. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 1970; 227:680–5.

22 Wang B, Lim DJ, Han J, Kim YS, Basbaum CB, Li JD. Novelcytoplasmic proteins of nontypeable Haemophilus influenzaeup-regulate human MUC5AC mucin transcription via a positivep38 mitogen-activated protein kinase pathway and a negativephosphoinositide 3-kinase-Akt pathway. J. Biol. Chem. 2002;277: 949–57.

23 Ihle JN, Nosaka T, Thierfelder W, Quelle FW, Shimoda K. Jaksand Stats in cytokine signaling. Stem Cells 1997; 15: 105–11.

24 Wery-Zennaro S, Zugaza JL, Letourneur M, Bertoglio J, Pierre J.IL-4 regulation of IL-6 production involves Rac/Cdc42- and p38MAPK-dependent pathways in keratinocytes. Oncogene 2000;19: 1596–604.

25 Li JD, Feng W, Gallup M et al. Activation of NF-κB via a Src-dependent Ras-MAPK-pp90rsk pathway is required for Pseudo-monas aeruginosa-induced mucin overproduction in epithelialcells. Proc. Natl Acad. Sci. USA 1998; 95: 5718–23.

26 Smirnova MG, Kiselev SL, Birchall JP, Pearson JP. Up-regulationof mucin secretion in HT29-MTX cells by the pro-inflammatorycytokines tumor necrosis factor-α and interleukin-6. Eur.Cytokine Netw. 2001; 12: 119–25.

27 Tang G, Minemoto Y, Dibling B et al. Inhibition of JNK activa-tion through NF-kappaB target genes. Nature 2001; 414:313–17.

28 Chantret I, Barbat A, Dussaulx E, Brattain MG, Zweibaum A.Epithelial polarity, villin expression, and enterocytic differentia-tion of cultured human colon carcinoma cells: a survey of twentycell lines. Cancer Res. 1988; 48: 1936–42.

29 Hollingsworth MA, Strawhecker JM, Caffrey TC, Mack DR.Expression of MUC1, MUC2, MUC3 and MUC4 mucin mRNAsin human pancreatic and intestinal tumor cell lines. Int. J. Cancer1994; 57: 198–203.

282 J Iwashita et al.

30 Lesuffleur T, Porchet N, Aubert JP et al. Differential expressionof the human mucin genes MUC1 to MUC5 in relation to growthand differentiation of different mucus-secreting HT-29 cellsubpopulations. J. Cell. Sci. 1993; 106: 771–83.

31 Grunig G, Warnock M, Wakil AE et al. Requirement for IL-13independently of IL-4 in experimental asthma. Science 1988;282: 2261–3.

32 Wills-Karp M, Luyimbazi J, Xu X et al. Interleukin-13: centralmediator of allergic asthma. Science 1998; 282: 2258–61.

33 Ordonez CL, Khashayar R, Wong HH et al. Mild and moderateasthma is associated with airway goblet cell hyperplasia andabnormalities in mucin gene expression. Am. J. Respir. Crit.Care. Med. 2001; 163: 517–23.