Cytokine modulation of intestinal epithelial cell restitution: Central role of transforming growth factor β

Cytokine Wlodulation of Intestinal Epithelial Cell Restltution: Central Role of Transforming Growth Factor p AXEL U. DIGNASS and DANIEL K. PODOLSKY Gastrointestinal Unit, Department of Medicine, and Massachusetts General Hospital/NERPRC Center for the Study of Inflammatory Bowel Disease, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts

Bac&round: After various forms of superficial injury, mucosal integrlty is re-established by rapid migration of epithelial cells across the wound margins in a process termed restitution. The aim of the present study was to assess the role of several regulatory peptides produced wlthln the intestinal mucosa in epithelial restitution. Methods: The effects of various cytokines and peptlde growth factors were studied in an in vitro model of intestinal epithelial restitution. Standard “wounds” were established in confluent monolayers of the intestlnal cell line IEC-6, and migration was quantitated in the presence or absence of the physiologically relevant cytokines transforming growth factor (TGF)-a, epidermal growth factor (EGF), interleukin (IL)-18, IL-6, tumor necrosis factor (TNF)-a, interferon gamma (K-N-y), and platelet-derived growth factor (PDGF). Results: Four factors (TGF-a, EGF, IL-lb, and IFN-y) enhanced epitheliai cell restitution by 2.3-fold to 55fold. In contrast, IL-6, TNF-a, PDGF,and an endotoxin llpopolysaccharide had no effect on cell migration. Enhancement of restitution was independent of proliferation. The restitution-promoting cytokines TGFa, EGF, IL-lb, and IFN-y increase the production of bioactive TGF-fI1peptide in wounded IEC-6 cell monolayer. The promotion of IEC-6 restitution by various cytokines coukl be completely blocked by addition of immunoneutralizlng anti-TGF-bl . Conclusions: These findings suggest that various cytokines that are expressed in intestinal mucosa promote epithelial restitution after mucosal injury through increased productlon of bioactlve TGF-Bl in epithelial cells.

T

he mucosal epithelium of the alimentary tract forms a barrier to the broad spectrum of noxious substances present within the lumen. Rapid resealing of this barrier after injuries is essential to preservation of normal homeostasis. Observations made over the past several years have shown the capability of the gastrointestinal tract to rapidly re-establish continuity of the surface epithelium after extensive destruction. The process by which epithelial continuity is rapidly re-established after various forms of injuries has been termed epithelial restitution and occurs both in vitro

and in vivo.‘-6 Restitution occurs by migration of viable epithelial cells from areas adjacent to or just beneath the injured surface to cover the denuded area. This process does not require cell proliferation. In vivo, restitution has been observed to occur within minutes to hours. The rapidity of this process suggests that cellular proliferation is not essential to restitution; regeneration through epithelial cell proliferation and differentiation of surface mucosal cells requires at least 1-2 days.‘-” Previous studies in this laboratory using an in vitro model of epithelial restitution have shown that transforming growth factor (TGF)-P promotes intestinal epithelial restitution. ‘OHowever, a broad spectrum of regulatory peptides generally known as cytokines, which have recently been identified within the intestinal mucosa, may be important modulators of epithelial cell function. ” These regulatory peptides encompass structurally diverse peptides identified variously as peptide growth factors, interleukins, interferons, and colony-stimulating factors. Several cytokines have been found to be expressed by constituents of the intestinal mucosa and modulate intestinal epithelial cell populations. They include TGF-P, TGF-CZ, interleukin (IL)-1, tumor necrosis factor (TNF), and interferongamma (IFN-)I). l*-” These cytokines also exert potent effects on constituents of the mucosal immune system. A high degree of redundancy is present within the “network” of cytokines found in the intestinal mucosa; a single cytokine can be produced by a wide variety of cells, and it acts on a variety of distinct cell populations with diverse bioactivities.” Both epithelial and immune cell populations may produce many Abbreviations used in this paper: DMEM, Dulbecco’s modlfied Eagle’s medium; EGF,epidermal growth factor; FCS, fetal calf serum; GAPD, glyceraldehyde-3-phosphate dehydrogenase; H-T, Interferon y; IL, interleukin; LPS, Ilpopotysaccharide; MvlLu, mink lung eplthellal cells; PDGF,platelet-derived growth factor; SSC, standard saline citrate; TGF,transforming growth factor; TNF,tumor necrosis factor. 0 1993 by the American Gastroenterokgkal Association 0016-5085/93/$3.00

1324

DIGNASS

AND

PODOLSKY

or all of these cytokines in addition to serving as target cells. These observations lend support to the notion that a complex network of cytokine interactions within the mucosal system of the gastrointestinal tract may have an important role in the regulation of a range of biological processes within the intestinal mucosa, e.g., regulation of epithelial cell proliferation and differentiation. A number of observations have suggested that cytokines also play a central role in unscheduled remodeling of tissues. ‘i The process of unscheduled remodeling of tissues occurs after injuries, inflammation, and infection of the intestine caused by various mechanisms (e.g., infectious agents, toxins, radiation, immunologic disorders, etc.). As noted above, the process of epithelial restitution is a central process in host mechanisms to preserve or re-establish the mucosal barrier. Recent reports have shown that early restitution of superficial mucosal injuries in the gastrointestinal tract requires migration of epithelial cells; this migration can be modulated by some cytokines as assessed in in vitro models of intestinal epithelial cell restitution 6,10,20-24 In the present report, we explore the role of several cytokines in epithelial restitution using an in vitro wounding model of confluent monolayers of the rat intestinal epithelial cell line IEC-6. Furthermore, we describe studies to characterize the mechanisms of action by which these cytokines and those previously known to enhance epithelial cell migration modulate epithelial cell restitution. These cytokines are physiologically produced within the intestinal mucosa or have target cells within the constituents of the intestinal mucosa.

Materials and Methods Materials Radiochemicals were obtained from New England Nuclear, Boston, MA. Human recombinant TGF-U, porcine platelet TGF-PI, and human recombinant IL-Ip were obtained from R&D Systems, Minneapolis, MN, recombinant human IL-6 from Boehringer Mannheim, Indianapolis, IN; recombinant mouse TNF-a from Genzyme, Cambridge, MA; recombinant rat IFN-y from Gibco, Gaithersburg, MD, turkey antihuman TGF-fi and human platelet-derived growth factor (PDGF) from Collaborative Research, Bedford, MA, and mouse submaxillary epidermal growth factor (EGF) from BTI, Stoughton, MA. Lipopolysaccharide (LPS) from E&+&a co/i serotype 055:B5 was obtained from Sigma, St. Louis, MO. A polyclonal rabbit antibody (A1/29) raised against a synthetic peptide of TGFPI and crossreacting with mature TGF-Pz5 was kindly provided as a gift from Celtrix Laboratories, Palo Alto, CA, and

GASTROENTEROLOGY

Vol. 105,

No. 5

used for immunohistochemical studies. IEC-6 cells first established in this laboratory by Dr. Andrea Quaroni were used at 15th-18th passage. 26 Mink lung epithelial cells (MvILu) cells were obtained from American Type Culture Collection (Rockville, MD) and used at 47th-53rd passage.

Wounding Assays Wound assays were essentially performed as previously described” using a modified method described by Sato and Rifkin.27 Confluent monolayers of IEC-6 cells in 60mm plastic dishes were wounded with a razor blade; generally, two wounds approximately 20-25 mm across the dish were made and separated by about 1.5 cm. Cells were washed with fresh serum-deprived medium, and the wounded monolayers were further cultured for 24 hours in fresh serum-deprived medium in the presence or absence of individual growth factors and cytokines TGF-a (lo-250 ng/mL), EGF (lo-250 ng/mL), IL-Ip (IO-1000 pg/mL), IFN-)I (100-1000 U/mL), IL-6 (500-1000 U/mL), TNF-a (100-1000 U/mL), and PDGF (0.5-5.0 U/mL). Migration of IEC-6 cells was assessed in a blinded fashion to avoid observer bias by determination of the number of IEC-6 cells observed across the wound border expressed as the mean number of cells found across the wound border in a standardized wound area. Wound areas were standardized by taking photomicrographs at lOO-fold magnification using an inverted microscope Nikon Diaphor TMS and a Nikon N6006 camera (Nikon Inc., Melville, NY). Experiments were performed in triplicate, and several wound areas per plate were used to quantitate migration. Data presented under results are expressed as mean value f SD of at least three independent experiments. Wound assays as previously described were also performed after mitomycin C pretreatment of confluent IEC-6 monolayers to distinguish between enhanced restitution by increased migration or proliferation. Preliminary studies using [3H]thymidine uptake assays showed that a dose of 2 /tg/mL mitomycin C inhibited proliferation entirely without affecting the viability of IEC-6 cells. Statistical analysis of data was performed using Student’s f test for paired data.

Determination

of TGF-Pl

Messenger

RNA

Total cellular RNA was isolated from IEC-6 monolayers (intact or wounded) by modification of the method of Chirgwin et al. as described. ‘* Messenger RNA (mRNA) (4 pg/lane) obtained by oligo(dT)-cellulose column chromatography was electrophoresed in a 1.0% formaldehyde agarose gel and blotted onto Nylon Transfer Membranes (MSI, Westboro, MA) by standard methods. Human TGF-PI and human glyceraldehyde-3-phosphate dehydrogenase (GAPD) probes were prepared respectively by random priming of a 2.14-kilobase (kb) EtoRI insert of a human TGF-PI complementary DNA (cDNA) clone designated pHTGF-P2 and a 0.78-kb P~t1, Xba I insert of a human GAPD plasmid designated pHcGAP obtained from American Type Culture Collection. Hybridization was performed at 42°C in 50% form-

MODULATION OF EPITHELIAL RESTITUTION BY CYTOKINES

November 1993

amide, 5X standard saline citrate (SSC), 5X Denhardt’s solution, and 10% dextran sulfate; blots were successively washed at 50°C in 2X SSC and 0.1% sodium dodecylsulfate three times. Blots were first hybridized with the TGF-PI probe and exposed for 16-48 hours at -80°C. The same blots were then hybridized with GAPD probe to standardize mRNA loading as previously described.29*m Relative abundance of TGF-P1 transcript was assessed by laser densitometric scanning normalized to the density of GAPD transcript.

Determination TGF-j.3

vine serum albumin) for 28 hours at 4°C in a humidified chamber. Bound primary antibody was localized with the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) using diaminobenzidinetetrahydrochloride as a substrate according to a protocol suggested by the manufacturer. Three controls were selected for the experiments. Wounded monolayers were incubated with PBS or rabbit preimmune serum instead of TGF-Pl antibody. In addition, the incubation with anti-TGF-P antibody was performed in the presence of excess porcine TGF-Pl.

of Latent and Bioactive

Latent and bioactive TGF-P were determined by a bioassay using Mvl Lu as described elsewhere.3’ Briefly, samples were activated by acidification with 150 mmol/L HCl for 60 minutes followed by neutralization before assay for TGF-P bioactivity. Subconfluent MvlLu cells were used to initiate the growth inhibition assay. Then 1 X lo5 cells in 0.2% fetal calf serum/Dulbecco’s modified Eagle’s medium (FCS/DMEM) per well were seeded on 24-well plates. Five hours after seeding, test samples and standardized quantities of porcine platelet TGF-PI (R&D Systems) were added and cells were incubated for 20 hours at 37’C. [3H]Thymidine (1.5 pCi/well) was added, and the incubation was continued for an additional 4-hour period. The cells were washed with phosphate-buffered saline (PBS) and fixed in methanol:acetic acid (3:1, vol/vol). Acid insoluble material was then lysed with 1N NaOH, and radioactivity was counted using a liquid scintillation counter.

Determination in IEC-6 Cells

of Wounded

IEC-6

Statistical

of IEC-6 Cell

Wounded monolayers were washed with PBS and cells fixed in 50% acetone-50% methanol. Endogenous peroxidase activity was blocked by incubation with 0.3% HaO,. Monolayers were incubated with the primary polyclonal rabbit anti-TGF-Bl antiserum (20 pg/mL PBS and 3% bo-

Incorporation

Analysis

Results are expressed as mean f SD of the mean. Statistical significance of differences between mean values was assessed with the Student’s t test for paired and unpaired data. All reported significance levels represent two-tailed P values.

Results tween

studies

have shown

the untransformed

thelial

cell. 17-19,26Im portant

cell line and the primary be assumed terocyte; model

epithelial

nonetheless,

between

counterpart capacity

into a phenotypically

intestinal

portunity

differences

be-

rat epithelial

rat crypt intestinal

in view of the limited

differentiation

the similarity

small intestinal

IEC-6 cell line and the normal

IEC-6 cells were grown to confluence on plastic tissue culture chambers (Lab Tek Chamber Slide; Nunc Inc., Naperville, IL). Wounding assays were performed essentially as described above. Following wounding, the medium was changed to DMEM containing 0.1% FCS supplemented with various cytokines. After an incubation period of 20 hours, [3H]thymidine (10 pCi/plate) was added, and incubation was continued for an additional 4-hour period. After washing the wounded monolayers with PBS, cells were fixed in 100% methanol. Slides were then coated with autoradiography emulsion NTB II (Eastman Kodak, Rochester, NY) and exposed for 3-7 days at 4°C. Slides were developed using Developer D-19 and Kodafix solution (Eastman Kodak) as suggested by the manufacturer.

lmmunohistochemistry Monolayers

of Thymidine

Subconfluent IEC-6 cells maintained in 24-well multiwell plates were changed to fresh DMEM containing 0.1% FCS 17 hours before addition of cytokines. [3H]Thymidine (1.8 pCi/well) was added after a 20-hour incubation period with cytokines, and incorporation was determined after further incubation for 4 hours by standard techniques. lo

Extensive

Autoradiography Monolayers

1325

epithis

have to

for terminal

mature

viilus

en-

the IEC-6 cell line provides

epithelial

to study epithelial

a

cell line that offers the opcell responses

without

the

ambiguity of contamination with nonepithelial constituents inherent to studies with preparations of primary intestinal

epithelial

Addition TGF-ol

cells.

of the peptide

(lo-250

ng/mL),

growth EGF

factors/cytokines

(lo-250

ng/mL),

IL-

lp (10-1000pg/mL), and IFN-y (10-1000 U/mL) serum-deprived medium enhanced the migration IEC-6 cells into the denuded in a dose-dependent manner

to of

area of a model “wound” compared with migration

of IEC-6 cells cultured in serum-deprived medium alone (Figure 1). Migration of cells into the denuded wound area can be observed as early as 4 hours after producing the wound in the presence of restitutionenhancing cytokines. To obtain reliable numbers of

1326

DIGNASS AND F’DDOLSKY

10

0

1

loo

50

GASTROENTEROLOGY Vol. 105. No. 5

0

250

10

TGFa [n&d]

100

Jo

0

250

10

500

100

IWO

IL1 B Wml

EGF [tag/ml]

Figure 1. Dose-dependent

0

1

10

100

500

effects of cytokines on migration of IEC-6 cells. Confluent monolayers maintained for 24 hours in DMEM containing 0.1% FCS were wounded and then cuitured for 24 hours after replacement of medium with fresh DMEM-0.1% FCS in the presence of the following cytokines: (A) TGF-a (10,50, 100, and 250 ng/mL), (5) EGF (10,50, 100, and 250 ng/mL), (C) IL- 16 (10, 100, 500, and 1000 pg/mL), and (D) IFN-T (10, 100, 500, and 1000 U/mL). Migratidn of IEC-6 cells was quantitated as.detailed.under Materials and Methods.

1004

IFN-y ~/ml]

cells across the wound

edge, a standardized

incubation

time of 24 hours was used for all experiments. Having shown a dose-dependent stimulation of epithelial cell migration IFN-)I),

by four cytokines further

concentrations thelial

studies

(TGF-a,

were

that caused maximal

cell migration.

As shown

EGF,

performed

IL-1 p, and at cytokine

stimulation

in Figure

2/l,

of epimigra-

of enhanced

IEC-6

cell migration,

rial endotoxins

(LB,

was assessed.

As shown

affect IEC-6

cell migration.

Similar

stimulation

served when layers

after

cytokines pretreatment

0.1-1.0

the effect of bacte-

pg/mL)

in Figure

of IEC-6

2A,

on migration LPS did not

cell migration

were added to wounded with

was obmono-

replication-inhibiting

tion was significantly (p < 0.005) enhanced by an average of four-fold by TGF-a, 4.6fold by EGF (100 ng/mL), 2.9-fold by IL-1p (500 pg/mL), and 3.8-fold by IFN-y (1000 U/mL). No significant effect on mi-

doses of mitomycin C (2 pg/mL tion of IEC-6 cell proliferation

for 2 hours). by mitomycin

InhibiC was

gration was observed when IL-6 (500-1000 U/mL), TNF-CY, (100-1000 U/mL), or PDGF (0.5-5.0 U/mL) were added to media after wounding as indicated in Figure 2A. The latter findings suggest that the stimulation of cell migration after addition of TGF-a, EGF, IL- 1 p, or IFN-y was specific. To exclude the possibility that endotoxins (which may contaminate recombinant growth factor preparations) are the driving force

IEC-6 cells with mitomycin C (2 pg/mL) inhibited cell proliferation as assessed by thymidine incorporation of stimulated (TGF-a) and quiescent IEC-6 cells, even below the level of inhibition observed in the presence of the potent proliferation-inhibiting growth factor TGF-Pl. Enhanced restitution after previous treatment with mitomycin C suggests that the effect of is EGF, TGF-cr, IL-1 p, and IFN-y on IEC-6 migration

confirmed by [3H]thymidine incorporation assays. The data presented in Table 1 indicate that pretreatment of

November 1993

MODULATION OF EPITHELIAL RESTITUTION BY CYTOKINES

1327

OTOFU

own

IL-1 p

IFN-7

PDCP

Figure 2. Effect of cytokines on migration of IEC-6 cells. (A) Confluent monolayers maintained for 24 hours in DMEM containing 0.1% FCS were wounded and then cultured for 24 hours after replacement of medium with fresh DMEM 0.1% FCS in the presence of the following cytokines: TGF-a (100 ng/mL), EGF (100 ng/mL), IL- 1p (500 pg/ mL), IFN-T (1000 U/mL), IL-6 (500 U/mL), TNF-a (1000 U/mL), and PDGF (2.5 U/mL). Migration of IEC-6 cells was determined as detailed under Materials and Methods. *P 5 0.005 vs. control (DMEM). (6) Confluent monolayers maintained for 24 hours in DMEM containing 0.1% FCS were exposed for 2 hours to mitomycin C (2 pg/mL). The monolayers were washed extensively and subsequently wounded and treated with cytokines as described for A. *P 5 0.01 vs. control.

Figure 3. Autoradiographs of wounded IEC-6 monolayers. Confluent monolayers of IEC-6 cells were changed to DMEM containing 0.1% FCS. After 24 hours, l-cm wide wounds were made with a razor blade as described in Materials and Methods. Media was changed to fresh DMEM and 0.1% FCS alone or supplemented with cytokines. After 20 hours, [3H]thymidine (10 @/plate) was added and incubation continued for an additional 4 hours. Slides were then exposed to autoradiography emulsion and developed as described in Materials and Methods. IEC-6 monolayers are counterstained with H&E. (A) Control; DMEM containing 0.1% FCS. (13)Treatment with TGF-a (100 ng/mL). (C) Treatment with IL-1 S (500 pg/mL). (D) Treatment with IFN-T (1000 U/mL).

migrating independent

of

proliferation

proof that enhanced eration wounded proliferating liferating poration.

restitution

was obtained IEC-6

(Figure

cell monolayers.

cells reflecting As shown

Using

analysis

of

this method,

by black nuclei

radiolabeled

in Figure

Further

is not caused by prolif-

by autoradiographic

cells are identified

2B).

thymidine

3, few nuclei

in proincorof cells

served

DMEM-0.1% FCS t TGF-a (100 ng/mL) t TGF-/3 (5.0 ng/mL)

No mitomycin

Mitomycin C

18,474 I~I986 25,932 rt 3658 5331+ 149

1656T 110 2128 + 77 1919 k 166

NOTE. All values are expressed as counts per minute + SEM. Semiconfluent IEC-6 cells were maintained in 0.1% FCS DMEM in 24-well plates. Cells were incubated either with or without mitomycin C (2 pg,/mL) for 2 hours before addition of growth factors or control medium (DMEM 0.1% FCS). Thymidine incorporation was assessed as described in Materials and Methods.

the wound

in control-

after wounding, tion studies in Table

edge show

label, and no significant

To quantify

or cytokine-treated

the thymidine IEC-6

2, no significant

in IEC-6

thymidine

cells

incorpora-

monolayers.

changes

of the are ob-

monolayers.

uptake

we performed

of wounded

uptake

differences

As shown

in thymidine

incor-

poration are observed in unwounded and wounded monolayers of IEC-6 cells cultured in the presence of cytokines

Table 1. Effect of Mitomycin C on Thymidine Incorporation in IEC-6 Cells

across

radioactive

mary,

or serum-deprived

these results

restitution

indicate

is a result

medium

alone.

that enhanced

of increased

In sum-

IEC-6

cell migration

cell and

not cell proliferation. These results are in accordance with our previous data” that enhanced migration of IEC-6 cells observed in the presence of exogenous TGF-P was independent of cellular proliferation. Previous studies” had shown that the expression of TGF-P mRNA was significantly enhanced in wounded monolayers compared with unwounded controls. Increased migration of IEC-6 cells was observed in the presence of exogenous TGF-P1 as well as with condi-

1328

GASTROENTEROLOGY Vol. 105, No. 5

DIGNASS AND PDDDLSKY

Table 2. Incorporation of Thymidine in Wounded and Unwounded IEC-6 Cell Monolayers Unwounded DMEM 0.1% FCS + EGF (100 ng/mL) + TGF-(U (100 ng/mL) + IL- 1B (500 pg/mL)

5 1,983 86,278 79,569 49,606

f + f +

5607 867 3234 9327

Wounded 56,370 81,084 72,761 56,931

f 4453 + 3314 z!z 11,196 + 3684

NOTE. All values are expressed as counts per minute f SEM. Confluent IEC-6 cells. were maintained in 0.1% FCS DMEM in 6-well plates. One wound (15 mm X 5 mm) per well was established. After several washes with serum-deprived medium of wounded and unwounded monolayers, wounded and unwounded monolayers were cultured for 24 hours in fresh serum-deprived medium in the presence or absence of individual growth factors as indicated. [3H]Thymidine (4 pCi/well) was added for the last 4 hours of the 24-hour incubation period, and thymidine incorporation of IEC-6 cells was measured as outlined in Materials and Methods.

tioned medium from wounded IEC-6 cell monolayers. Enhancement of cell migration could be blocked entirely by addition of immunoneutralizing TGF-P antisera, indicating that TGF-P is the dominant intrinsic factor contributing to enhanced migration in the wounded monolayers. lo To assess the role of TGF-j3 as a possible m e d ia t or of enhanced epithelial restitution observed in the presence of several cytokines, the concentration of TGF-P peptide in the culture media of IEC-6 cells after a 24-hour incubation period following wounding was assessed. As Figure 4 shows, the concentration of bioactive TGF-P was increased in media obtained from wounded IEC-6 cells cultured in the presence of TGF-a/EGF (100 ng/mL), IFN-)I (1000 U/mL), and IL-lb (500 pg/mL) compared with control media of wounded IEC-6 cells cultured in serum-deprived medium alone (P < 0.03). No statistically significant changes were observed for latent TGF-P. Immunohistochemical staining of wounded monolayers using an antibody that cross-reacts with mature TGF-p125 showed increased staining activity as early as 2 hours after establishing the wound; strongest staining was observed 12 hours after establishing the wound. Enhanced staining levels persisted 24 hours after wounding. No staining was observed in various control situations described in Materials and Methods (Figure 5). Immunohistochemical staining for mature TGF-Pl revealed enhanced production of bioactive TGF-P 1 in wounded monolayers treated with TGF-CX, IL-lj3, or IFN-)I compared with wounded monolayer controls in serum-deprived medium alone (Figure 5). Interestingly, a diffuse staining pattern with peptide present in patches was observed; notably, increased production

of bioactive TGF-PI peptide was not selectively enhanced in proximity to the wound, but scattered throughout the residual monolayers. To further assess the role of TGF-j3 as a mediator of cytokine-enhanced epithelial restitution in wounded IEC-6 monolayers, the expression of TGF-j31 at the mRNA level was evaluated in wounded and unwounded monolayers. As shown in Figure 6, wounding of confluent monolayers led to significantly increased steady-state expression of TGF-Pl mRNA compared with unwounded controls (P < 0.04 = P < 0.001). Furthermore, the absolute increase of TGF-Pl mRNA expression in wounded monolayers in the presence of EGF/TGF-a and IL-lp but not enhanced IFN-y is more pronounced but not statistically significant compared with monolayers cultured in serum-deprived medium alone. Interestingly, IL-1 p causes a significant induction of TGF-P1 mRNA in confluent wounded and unwounded monolayers of IEC-6 cells compared with monolayers cultured in serum-deprived medium alone. To assess the functional relevance of increased levels of bioactive TGF-P peptide in wounded IEC-6 monolayers after treatment with several cytokines, the migration of IEC-6 cells in wounded monolayers was determined in the presence of immunoneutralizing antiTGF-P (27.5 &/mL) added to the medium at the time of wounding. Preliminary studies had shown the

6-

7bb4TGF-p [nglml]

3zk _ lO-

1 IL-l@

+IFNy

Figure 4. Effect of cytokines and wounding on production of latent and bioactive TGF-6 by IEC-6 monolayers. Confluent monolayer of IEC-6 cells maintained for 24 hours in DMEM containing 0.1% FCS were wounded and then cultured for 24 hours after replacement of medium with fresh DMEM containing 0.1% FCS supplemented with cytokines as indicated (for concentrations, see Figure 2A and text). Media were collected after 24 hours, and latent and bioactive TGF-5 was determined as outlined in Materials and Methods. Columns with error bars represent mean value of three independent experiments ? SD of the mean. *P < 0.03 vs. control (DMEM bioactive). 0, latent; 0, bioactive.

MODULATION OF EPITHELIAL RESTITUTION BY CYTOKINES

November 1993

Figure 5. lmmunohistochemical staining for TGF-6, of wounded IEC-6 monolayer. Confluent monolayers of IEC-6 cells were wounded as described in Materials and Methods, and the medium was replaced with fresh DMEM and 0.1% FCS supplemented with cytokines. After a 24-hour incubation period, monolayers were fixed, and immunohistochemical staining for TGF-0, was performed as outlined in Materials and Methods. (A) Normal preimmune serum, control. No staining was observed in other control situations described in Materials and Methods (data not shown). (13) Incubation with DMEM 0.1% FCS. (C) TGF-a treated (100 ng/mL). (D) IL- 16 treated (500 pg/mL).

ability

of this

nous TGF-P1

antibody

by [3H]thymidine Furthermore, levels

observed

when

antiTGF-pl

medium.”

This

munoneutralization into

uptake

we have

cell restitution

the medium

could

exoge-

wound

as assessed

termed

to immunoneutralize

added to the culture

medium

assays and wounding previously

shown

be reduced

even below

for serum-deprived antibody observation

medium

was added

cells.

TGF-P

utes and appears

IEC-6

Proliferation

mucosal

Restitution

to be independent

of mucosal

min-

of proliferation.

‘T’,~

cells to increase

the

controls by imin

by sev(27.5

DCscussion Mucosal healing after superficial injury requires migration of epithelial cells adjacent to the

mucosal several lating

epithelial peptide mucosal

available

a process within

occurs in a more delayed fashion, generally beginning hours or days after the injury.’ The proliferation of

secreted

of enterocytes

epithelial

defect, begins

pool

pg/mL), supporting the conclusion that the observed increase in IEC-6 cell migration was caused by enhanced production of bioactive TGF-P.

initial

the

baseline

As indicated

Figure 7, the stimulation of IEC-6 restitution eral cytokines could be blocked by anti-TGF-pl

across

assays.

to the culture

can be explained

of endogenous by epithelial

that

area

restitution.4,“s

to resurface

cells is known

growth restitution

of epithelial restitution culty in distinguishing

factors,

the defect

to be regulated but the factors

are less understood.

by regu-

Studies

in vivo are limited by the diffithe effects of cell migration

from the delayed phase of mucosal healing. To overcome the difficulties inherent to in vivo studies, an in vitro model of epithelial cell injury has been used to study mechanisms of epithelial restitution. This model uses the untransformed rat intestinal epithelial cell line IEC-6. In contrast to primary epithelial cell preparations which may be contaminated by nonepithelial

1330

DIGNASS AND PODOLSKY

OMEM

n=o

TGFa

n-8

GASTROENTEROLOGY Vol. 105, No. 5

‘i-2

IFlnrr II=6

Figure 6. Effect of cytokines on the expression of TGF-/3,mRNA in unwounded (B) and wounded (0) IEC-6 monolayers. mRNA was prepared from confluent monolayers either intact or 24 hours after wounding. Monolayers were cultured in the presence or absence of cytokines as indicated in the figure. The relative expression of TGF-P, standardized to GAPD was quantitated by laser densitometry. Columns represent mean values + SD of the mean of five to eight Northern blots obtained in independent experiments and used for densitometric analysis. n indicates number of sample size.

of the intestinal mucosa, the IEC-6 cell line serves as a useful tool to study isolated intestinal epithelial cell responses without the ambiguity of primary epithelial cell preparations. Numerous growth factors and cytokines are expressed within the intestinal mucosa. In addition, the various cell populations within the intestinal mucosa exhibit binding sites for several physiological growth factors and cytokines. Previous studies in this laboratory” showed that exogenous and endogenous TGF-P paradoxically promote intestinal epithelial restitution in vitro despite potent inhibition of epithelial cell proliferation. Other cytokines in the intestinal mucosa, including TGF-U, and its functional analogue EGF, are potent stimulators of epithelial cell proliferation, but little is known about their role in epithelial cell restitution. The studies in this report show that several cytokines specifically promote intestinal epithelial cell restitution in vitro. TGF-ol, EGF, IL-lb, and IFN-7 significantly enhanced intestinal epithelial cell restitution in vitro, whereas IL-6, TNF-CX, and PDGF (cytokines that can also be found in the intestinal mucosa) had no effect on this process. All of the investigated cytokines have been shown to modulate other functional properties of constituents of the intestinal mucosa.‘“‘” Interestingly, TGF-CY,,EGF, IL-1 p, and IFNy enhance epithelial cell restitution independently of cell proliferation. The absence of an apparent effect on proliferation in this setting by TGF-(r. and EGF (which presumably utilize the same cell surface receptor) is especially noteworthy; these peptides have been constituents

observed to stimulate proliferation

of intestinal epithe-

lial cell lines in vitro and mucosal proliferation in vivo. The absence of a stimulatory effect may reflect increased levels of bioactive TGF-P1 found in the supernatants of wounded IEC-6 monolayers after cytokine treatment. The functional importance of this cytokine-mediated modulation of expression and production of bioactive TGF-P was shown by the ability of immunoneutralizing TGF-P antibody to block the cytokine-enhanced epithelial restitution. Collectively, these data suggest that TGF-P is a central factor intrinsically contributing to migration in the wounded monolayer and that several restitutionpromoting cytokines act through TGF-P. Furthermore, the presented data indicate that the cytokines enhance the bioactivation of TGF-P1 and to a lesser degree steady-state TGF-Pl mRNA expression. TGFp is known to be secreted as a biologically latent complex of the bioactive TGF-Pl dimer and the noncovalently linked proregion peptide. Approximately 90% of TGF-P released from platelets33 or TGF-P secreted into medium by cells in vitro34p35*36 is present in a latent form. The latent peptide propeptide complex is unable to bind to cellular receptors and is not recognized by antibodies to TGF-P.33 In vitro, the latent peptide can be “activated” (i.e., dissociated from its noncovalent association with the TGF-P propeptide and other proteins) by acidification, alkalinization, enzymatic digestion, or action of chaotropic agents.37 Since most cells have receptors for TGF-P,36 physiological control of the activation of the latent peptide provides an impor-

Figure 7. Effectof immunoneutralizingantiTGF-P on cytokine stimulation of IEC-6 cell migration. Confluent monolayers of IEC-6 cells were wounded as described in Materials and Methods. Medium was replaced by DMEM containing 0.1% FCSsupplemented with cytokines and monolayers were incubated for a 24-hour period in the presence or absence of antiTGFj3. Columns and error bars represent mean values + SD of the mean obtained from at least three independent experiments. *P i 0.005 vs. control (DMEM), **P < 0.005 vs. corresponding cytokine in the absence of antiTGF-P.

November

1993

mechanism to regulate its bioactivity. In previous studies,” we showed that wounded IEC-6 monolayers were competent to produce latent TGF-P as well as plasminlike protease activity needed to effect bioactivation through liberation of the mature TGF-P dimer. In the presence of protease inhibitors, the bioactivation of endogenously synthesized TGF-P could be blocked. The increased level of bioactive TGF-P in media of cytokine-treated wounded IEC-6 monolayers implies that these regulatory peptides increase the expression of both TGF-P as well as the bioactivating protease activities in wounded IEC-6 monolayers. Although the present studies suggest that several cytokines can play a role in the modulation of epithelial restitution, it should be noted that other factors may also contribute to enhanced epithelial restitution. Extracellular matrix components may be especially imhave shown alteration of portant. Recent studies 20,22,23 intestinal epithelial cell migration in vitro by the underlying matrix. Indeed, it has become evident in the past several years that the expression and functional activities of several growth factors and extracellular matrix factors are closely interrelated. It has been clearly shown that cell adherence to matrix can induce cells to produce cytokines.” Conversely, cytokines induce alterations in matrix production in many cells and also modulate integrin cell adhesion receptors.” These interactions provide a mechanism for the modulation of epithelial cell function in vitro and in vivo. In vivo, the effects of cytokines and extracellular matrix are further complicated by potential interactions with adjacent nonepithelial cells and inflammatory cells attracted to the wound area. Although the present studies exploit a simplified model to explore the role of cytokines in modulating epithelial restitution, further work will be needed to characterize the full complexity of the regulation of epithelial cell restitution in vivo. These findings indicate that several cytokines play a role in enhancing epithelial cell restitution following injury in vitro. These actions appear to be mediated through enhanced production of bioactive TGF-fi.

MODULATION OF EPITHELIAL RESTITUTION BY CYTOKINES

tant

References Morris GP, Wallace JL. The roles of ethanol and of acid in the production of gastric mucosal erosions in rats. Virchows Arch 13 Cell Pathol 198 1;38:23-38. Rutten MJ, Ito S. Morphology and electrophysiology of guinea pig gastric mucosal repair in vitro. Am J Physiol 1983;244G: 171182. Moore R, Carlson S. Madara JL. Rapid barrier restitution in an in vitro model of intestinal epithelial injury. Lab Invest 1989;60: 237-244. Feil W, Wentzl E, Vattay P, Starlinger M. Sogukoglu R, Schiessel

5.

6. 7.

8.

1331

R. Repair of rabbit duodenal mucosa after acid injury in vivo and in vitro. Gastroenterology 1987;92: 1973-1986. Waller DA, Thomas NW, Self TJ. Epithelial restitution in the large intestine of the rat following insult with bile salts. Virchows Arch A Pathol Anat Histopathol 1988;414:77-81. Nusrat A, Delp C, Madara J. Intestinal epithelial restitution. J Clin Invest 1992;89:1501-1511. Silen W. Gastric mucosal defense and repair. In: Johnson LR, ed. Physiology of the gastrointestinal tract. 2nd ed. New York: Raven, 1987:1044-1069. Lacy ER. Epithelial restitution in the gastrointestinal tract. J Clin

Gastroenterol 1988; 10 (Suppl 1):72-77. 9. Lipkin M, Sherlock P, Bell B. Cell proliferation kinetics in the gastrointestinal tract of man. II. Cell renewal in stomach, ileum, colon and rectum. Gastroenterology 1963;45:72 l-729. 10. Ciacci C, Lind SE, Podolsky DK. Transforming growth factor p regulation of migration in wounded rat intestinal epithelial monolayers. Gastroenterology 1993; 105:93- 10 1. in context. J Cell Biol 11. Nathan C, Sporn M. Cytokines

1991;113:981-986. 12. Shirota J, LeDug L, Yuan SY, Yothy S. lnterleukin 6 and its receptor are expressed in human intestinal epithelial cells. Virchows Arch B Cell Pathol 1990;58:303-308. 13. Madara JL, Stafford J. Interferon-gamma directly affects barrier function of cultured intestinal epithelial monolayers. J Clin Invest 1989;83:724-727. 14. Takacs L, Kovacs EJ, Smith MR, Young HA, Durum SK. Detection of IL- 1 alpha and IL- 1 beta gene expression by in situ hybridization. Tissue localization of IL-2 mRNA in the normal C57BL/6 mouse. J lmmunol 1988;141:3081-3095. 15. Wu SG, Miyamoto T. Radioprotection of the intestinal crypts of mice by recombinant human interleukin-la. Radiat Res

1990;123:112-115. 16. Chang EB, Musch MW, Mayer L. Interleukins 1 and 3 stimulate anion secretion in chicken intestine. Gastroenterology 1990;

98:1518-1524. 17. Suemori S, Ciacci C, Podolsky DK. Regulation of transforming growth factor expression in rat intestinal epithelial cell lines. J Clin Invest 1991;87;22 16-222 1. 18. Kurokawa M, Lynch K, Podolsky DK. Effects of growth factors on an intestinal epithelial cell line; transforming growth factor j3 inhibits proliferation and stimulates differentiation. Biochem Biophys Res Commun 1987;142:775-782. 19. Barnard JA, Beauchamp RD, Coffey RJ, Moses HL. Regulation of intestinal epithelial cell growth by transforming growth factor j3. Proc Natl Acad Sci USA 1989;86: 1578- 1582. 20. McCormack SA, Viar MJ, Johnson LR. Migration of IEC-6 cells: a model for mucosal healing. Am J Physiol 1992;263G:426-435. 21. Blay J, Brown KD. Epidermal growth factor promotes the chemotactic migration of cultured rat intestinal epithelial cells. J Cell Physiol 1985;24: 107- 112. 22. Basson MD, Modlin JM, Flynn SD, Jena BP, Madri JA. Independent modulation of enterocyte migration and proliferation by growth factors, matrix proteins, and pharmacologic agents in an in vitro model of mucosal healing. Surgery 1992; 1 12:299-308. 23. Basson MD, Modlin IM, Madri JA. Human enterocyte (Caco-2) migration is modulated in vitro by extra-cellular matrix composition and epidermal growth factor. J Clin Invest 1992;90: 15-23. 24. Kartha S, Toback FG. Adenine nucleotides stimulate migration in wounded cultures of kidney epithelial cells. J Clin Invest 1992;90:288-292. 25. Abbot BD, Birkbaum LS. TCDD-induced altered expression of growth factors may have a role in producing cleft palate and enhancing the incidence of clefts after coadministration of retinoic acid and TCDD. Toxicol Appl Pharmacol 1990;106:418-432. 26. Quaroni A, Wands J, Trelstad TL, lsselbacher KJ. Epithelial cell cultures from rat small intestine. J Cell Biol 1979;80:245-265.

1332

DIGNASS AND PODOLSKY

27. Sato Y, Rifkin DP. Inhibition of endothelial cell movement by pericytes and smooth muscle cells: activation of latent transforming growth factor B, like molecules by plasmin during coculture. J Cell Biol 1989; 109:309-3 15. 28. Chirgwin JM, Przybyla AF, Rutters WJ. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 1979; 185295-5299. 29. Scheving LA, Shurba RA, Nguyen TD, Gray GM. Epidermal growth factor receptor of the intestinal enterocyte. Localization to laterobasal but not brush border membrane. J Biol Chem 1989;264: 1735- 174 1. 30. Melton DA, Krieg DA, Rebagliati MR, Maniatis T, Zinn H, Green MR. Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage sp, promotor. Nucleic Acids Res 1984; 12:7055-7056. 31. Danielpour D, Dart LL, Flanders KC, Roberts AB, Sporn MB. Immunodetection and quantitation of the two forms of transforming growth factor p (TGF-6, and TGF-6,) secreted by cells in culture. J Cell Physiol 1989; 138:79-86. 32. Massague J. The transforming growth factor 6 family. Annu Rev Cell Biol 1990;6:597-64 1. 33. Pircher R, Jullien P, Lawrence DA. Beta transforming growth factor is stored in human blood platelets as a latent high molecular weight complex. Biochem Biophys Res Commun 1986; 136:3037.

GASTROENTEROLOGY Vol. 105, No. 5

34. Connor TB Jr, Roberts AB, Sporn MB, Danielpour D, Dart LL, Michels RG, de Bustros S, Enger C, Kato H, Lansing M, Hayashi H, Giaser BM. Correlation of fibrosis and TGFP type 2 levels in the eye. J Clin Invest 1989;83:1661-1666. 35. Lawrence DA, Pircher R, Kryceve-Martinerie C, Jullien P. Normal embryo fibroblasts release transforming growth factor in a latent form. J Cell Physiol 1984; 12 1:184- 188. 36. Wakefield LM, Smith DM, Matsui T, Harris CC, Sporn MB. Distribution and modulation of the cellular receptor for transforming growth factor 8. J Cell Biol 1987; 105:965-975. 37. Roberts AB, Sporn MB. The transforming growth factor betas, In: Sporn MB, Roberts AB, eds. Handbook of experimental pharmacology. Volume 95. Heidelberg, Germany: Springer-Verlag, 1990:419-472.

Received March 4, 1993. Accepted July 13, 1993. Address requests for reprints to: Daniel K. Podolsky, M.D., Gastrointestinal Unit, Jackson 7, Massachusetts General Hospital, 32 Frult Street, Boston, Massachusetts 02114. Supported by grants from the National Institutes of Health (DK 41557, DK 43351) and the Deutsche Forschungsgemelnschaft (Dl 477/1-l). The authors thank Yvonne J. Gllreath and Corollnda Helu for their help in the preparation of the manuscript.