APOPTOSIS is a morphologically and biochemically

Keratinocyte Growth Factor Decreases Total Parenteral Nutrition Induced Apoptosis in Mouse Intestinal Epithelium Via Bcl-2 By Barbara E. Wildhaber, Hua Yang, and Daniel H. Teitelbaum Ann Arbor, Michigan Background/Purpose: Total parenteral nutrition (TPN) induces epithelial cell (EC) apoptosis. Keratinocyte Growth Factor (KGF) increases EC-growth; however, little is known of its effect on apoptosis. This study aims to determine if mrna expression of Bcl-2 proteins (major mediators of epithelial cell apoptosis) is altered with TPN, and if KGF-administration influences Bcl-2 family expression. Methods: C57BL/6J mice (n 6 per group) received oral feeding (control), TPN (TPN), or TPN plus intravenous KGF daily (TPN KGF). After 7 days, intestine was harvested and EC isolated. Apoptosis was identified using flow cytometry. EC mrna expression of Bcl-2 family members was measured by reverse transcriptase polymerase chain reaction; Bcl-2 protein level was measured by immunoblot analysis. Results: EC apoptotic rates were: control, 14.4% 5.1%; TPN, 29.4% 11.3%; KGF, 17.2% 5.6%. Pro-apoptotic Bcl-2 proteins changed minimally with TPN or KGF; however, the antiapoptotic protein Bcl-2 changed significantly: control, 0.78 0.24; TPN, 0.10 0.13; KGF, 0.76 0.36. EC Bcl-2 protein levels were: control, 0.16 0.13; TPN 0.18 0.16; and TPN KGF 0.47 0.19. Conclusions: TPN-induced apoptosis decreased Bcl-2 mrna expression. KGF decreased EC apoptosis and increased Bcl-2 expression. Modalities to increase endogenous KGF, or KGFadministration may have benefit in patients on TPN. J Pediatr Surg 38:92-96. Copyright 2003, Elsevier Science (USA). All rights reserved. INDEX WORDS: Keratinocyte growth factor, intestinal epithelial cells, total parenteral nutrition, apoptosis, Bcl-2 protein family APOPTOSIS is a morphologically and biochemically distinct form of cell death, which plays an essential biological role in maintaining homeostasis and physiologic integrity in most tissues. Apoptosis of intestinal epithelial cells (EC) plays a critical role in total parenteral nutrition (TPN; absence of enteral nutrients) induced villus atrophy. 1 A major intracellular signaling pathway for apoptosis is mediated by the Bcl-2 family of proteins. Members of this protein family either function to prevent apoptosis (Bcl-2, Bcl-x L, Bcl-w) or support intracellular apoptotic signaling (Bax, Bak, Bid, Bad, Bcl-x S ). The up- or downregulation of these proteins play a major role in determining whether cells will undergo From the University of Michigan, Department of Pediatric Surgery, C.S. Mott Children s Hospital, Ann Arbor, MI. Presented at the 49th Annual Congress of the British Association of Paediatric Surgeons, Cambridge, England, July 23-26, 2002. This work was supported by a grant from the National Institute of Health, USA (AI44076-01) and by fellowships from the Stiftung Novartis Schweiz, the Swiss National Foundation, and the Swiss Society of Pediatric Surgery. Address reprint requests to Daniel H. Teitelbaum, MD, Section of Pediatric Surgery, University of Michigan Hospitals, Mott F3970, Box 0245, Ann Arbor, MI, 48109. Copyright 2003, Elsevier Science (USA). All rights reserved. 0022-3468/03/3801-0019$35.00/0 doi:10.1053/jpsu.2003.50018 apoptosis. 2 KGF is an EC-specific growth factor. In the gut, mucosal KGF is secreted by T-cell receptor intraepithelial lymphocytes (IEL). 3 We previously have shown that KGF expression in the mouse small intestine is downregulated with administration of TPN. 4 In previous studies, KGF has been used to increase EC growth, 5 however, little is known of its effect on apoptosis. The goal of the study was to determine if TPN-induced apoptosis of EC alters mrna expression of the Bcl-2 protein family. We also aimed to determine if KGF administration influences the rate of apoptosis and the expression of Bcl-2 proteins. Animals MATERIALS AND METHODS Male, specific pathogen-free, 2-month-old mice. C57BL/6J (Jackson Laboratories, Bar Harbor, ME) were anesthetized (sodium pentobarbital 50 mg/kg body weight, intraperitoneally), catheterized into the superior vena cava, and connected to an infusion pump (AIM pain provider pump, donated by Abbott Laboratories, Abbott Park, ILL). Dextrose 5% in 0.45 NS (0.9% saline) with 20 meq KCl/L was infused at an initial rate of 3 ml every 24 hours. TPN group. After 12 hours a TPN solution was started (7 ml every 24 hours), providing essentially the same caloric delivery among enterally fed mice. 6,7 No oral intake other than water ad libitum was permissible. 92 Journal of Pediatric Surgery, Vol 38, No 1 (January), 2003: pp 92-96

KGF DECREASES TPN-INDUCED INTESTINAL APOPTOSIS 93 Control group. In addition to the saline infusion (7 ml every 24 hours) mice received standard laboratory mouse chow and water ad libitum. TPN KGF group. After day 2, post-catheterization mice received daily intravenous KGF-injections (5 mg/kg/d rhukgf, Amgen, CA). Animals were killed after 7 days using CO 2, and the intestine was harvested. Mucosal Cell Isolation Isolation of mucosal cells was performed using a previously described protocol. 6 Flow Cytometry Apoptosis was determined by flow cytometer based on the cell surface expression of phosphatidylserine (Annexin V). Annexin V assay was performed with the Apoptosis Kit from BD PharMingen (San Diego, CA). Staining with propidium iodide was used to monitor cell necrosis. EC Purification Magnetic beads conjugated with antibody to CD45 (lymphocyte specific) were used to remove nonepithelial cells (BioMag SelectaPure Anti-Mouse CD 45R Antibody Particles, Polyscience Inc, Warrington, PA). Cells bound to beads were considered purified IEL; the supernatant containing EC was saved for RNA isolation. Flow cytometry confirmed purity of sorted cells, which was greater than 99%. Reverse Transcriptase Polymerase Chain Reaction Total RNA was isolated using Trizol (Gibco BRL, Gaithersburg, MD) and quantified by spectrophotometry at 260 nm absorption. EC mrna (poly-a positive) was reversed transcribed into cdna following a standard protocol. 6 Specific primers designed for Bcl-2, Bcl-x L, Bcl-w, Bax, Bak, Bid, Bad, Bcl-x S sequences were designed using an optimization program (OLIGO 4.1, National Biosciences, Plymouth, MN). The sequences for forward primers (FP) and reverse primers (RP) are as follows: Bcl-2 (gene bank accession number: AA867214) -FP: CAC ATC CAA TAA AAG AGC, Bcl-2-RP: ACC CCA TCC TGA AGA GTT; Bcl-x L (L35049) -FP: AGG CAG GCG ATG AGT TTG AAC, Bcl-x L -RP: GAA CCA CAC CAG CCA CAG TCA, Bcl-w (U59746) -FP: GTT TCC GCC GCA CCT TCT CT, Bcl-w-RP: CCC CGT CAG CAC TGT CCT CA; Bax (L22472) -FP: CGG CGA ATT GGA GAT GAA CTG, Bax-RP: GCA AAG TAG AAG AGG GCA ACC; Bak (Y13231) -FP: AAG ACG CTT TAG CAA ACA GG, Bak-RP: TAG GGA GGG CAA GGA TTG TG; Bid (U75506) -FP: GCC AAG CAC ATC ACA GAC CTG, Bid-RP: GAA GAC ATC ACG GAG CAA AGA T; Bad (L37296) -FP: GGA AGA CGC TAG TGC TAC AG, Bad-RP: GAG CCT CCT TTG CCC AAG TT; Bcl- x s (U10100) -FP: AGG CAG GCG ATG AGT TTG AAC, Bcl- x s -RP: GAA CCA CAC CAG CCA CAG TCA. PCR and gel were run under standard conditions. 6 To insure that DNA product was generated at the exponential portion of the product curve, for the various Bcl-2 family members the following cycle numbers were used: 28 cycles for Bak, Bad, Bid, Bcl-x L, Bcl-x S ; 40 cycles for Bcl-2, Bcl-w, Bax. Imaging was achieved digitally (Kodak EDAS System, Rochester, NY) and quantification with Kodak ID Image Analysis Software (Eastman Kodak). Results were expressed as the ratio of the investigated mrna over the -actin mrna expression. Immunoblot Analysis Western analysis was performed in a standard fashion with 20 g of total protein per sample. 8 First antibody was purified hamster antimouse Bcl-2 (1:1000, BD PharMingen, San Diego, CA). Second antibody was peroxidase-conjugated goat anti-armenian hamster IgG (1:2000, Jackson ImmunoResearch Laboratories, West Grove, PA). Equal loading was confirmed by blotting with an antibody to -actin. The same equipment as for PCR imaging was used for protein quantification. Results were expressed as the ratio of the Bcl-2 over the -actin protein expression. Statistical Analysis All results are expressed as mean SD. The results were analyzed using analysis of variance (ANOVA) and statistical significance was defined as P.05. RESULTS Apoptosis Apoptosis in EC in the small bowel significantly increased after 7 days of TPN: TPN, 29.4% 11.3% versus control, 14.4% 5.1% (P.02). KGF reversed this increase in apoptosis (17.2% 5.6%; P.03 v TPN; P.41 v control; Fig 1). In contrast, the percent of EC showing cell necrosis was not statistically different between control (6.8% 3.1%), TPN (15.4% 13.0%), and TPN KGF (7.4% 5.8%) mice (P.19 TPN v control; P.17 TPN KGF v TPN). Expression of Bcl-2 Protein Family mrna From control to TPN, Bak declined from 2.41 0.47 to 1.19 0.46 (minus 53%; P.01); Bad declined from 1.42 0.16 to 0.87 0.30 (minus 39%; P.01); Bcl-2 declined from 0.78 0.24 to 0.10 0.13 (minus 87%; P.01); and Bcl-w declined from 0.74 0.23 to 0.61 0.23 (minus 18%; P.28). Compared with TPN, administration of KGF increased Bak to 1.29 0.45 (by 8%; P.72); Bad increased to 0.99 0.33 (by 13%; P.45); Bcl-2 increased to 0.76 0.36 (by 660%; P.01); and Bcl-w increased to 0.88 0.19 (by 44%; P.04; Fig 2). The mrna expression of Bax, Bid, Bcl-x S, and Bcl-x L were not significantly different between control, TPN, and TPN KGF groups. Expression of Bcl-2 Protein Because the greatest change in the Bcl-2 family was Bcl-2 itself, the protein expression of Bcl-2 then was performed. Control (0.16 0.13) and TPN (0.18 0.16) levels were unchanged. After KGF administration, Bcl-2 protein expression was markedly increased: TPN KGF, 0.47 0.19 (P.05). DISCUSSION The intestinal mucosa has a rapid turnover rate (3 to 6 days) and works to maintain equilibrium between cell proliferation and cell death. A disregulation in the rate of apoptosis may result in tissue atrophy or in hyperplasia and malignant transformation. 9 The regulation of gastro-

94 WILDHABER, YANG, AND TEITELBAUM Fig 1. Percent of EC apoptosis after administration of TPN and TPN KGF. Apoptotic rate is detected by gating on EC with flow cytometry and detecting the percent of Annexin V staining (Y axis). Apoptotic rate is expressed as percentage of total EC. Note the increase in EC apoptosis in mice receiving TPN (29.4% 11.3%) versus control (14.4% 5.1%). Note the decline in apoptosis with the administration of KGF (17.2 5.6%). intestinal growth and differentiation is complex and induced by many factors. 10 Among those, KGF is a potent mitogen stimulating the proliferation and differentiation of epithelial cells (EC). 11 In the intestinal mucosa, KGF is secreted by activated T-cell receptor intraepithelial lymphocytes (IEL). 3 We have shown previously that with TPN administration, the IEL-derived KGF expression is downregulated. 4 Previous reports have used KGF to investigate the increase in EC growth. 5 In fact, Goodlad et al 12 has shown that KGF administration in a rat model of TPN resulted in prevention of villus atrophy. 12 However, little is known of its effect on apoptosis. A recent study found that KGF protects mouse hepatocytes against apoptosis. Hepatocytes pretreated with KGF exhibited reduced cell damage and DNA fragmentation. 13 In our mouse TPN model, the EC proliferation rate Fig 2. mrna expression of Bcl-2 family members. Note the distinct downregulation of Bcl-2 in TPN (gray) compared with controls (black), and the upregulation in the TPN KGF (white) bars. Results are expressed as a ratio to -actin mrna expression, mean SD. *P <.05. significantly decreased, and the apoptotic rate significantly increased. Application of KGF to TPN mice reversed the decline of proliferation of EC during TPN, and prevented the increase in EC apoptosis with TPN alone. The apoptotic rate of the EC nearly reached levels of control mice. These changes in the equilibrium of proliferation and apoptosis in EC are reflected in a significant decrease in villus height. Mechanisms of apoptosis vary among different cell types and may involve an array of signaling processes and regulators influencing multiple cellular functions. Generally, apoptosis relies on the activation of cysteine proteases of the caspase family. 14 The extrinsic apoptotic pathway is mediated by cell surface receptors and, in the intestinal tract, includes ligands on the IEL. Binding of the IEL s CD95 ligand (FasL) to the CD95 receptor on the EC activates the death-inducing signaling complex. Using the intrinsic pathway at the mitochondria-related level the engagement of apoptosis is regulated by the Bcl-2 protein family, which sets a survival threshold. These integral membrane proteins are either proapoptotic (Bax, Bak, Bid, Bad, Bcl-x S ), or antiapoptotic (Bcl-2, Bcl-x L, Bcl-w), and are localized to the nuclear membrane, endoplasmatic reticulum, and outer mitochondrial membranes. 15 Bcl-2 members may dictate the fate of the cell after receiving a death signal. We noted, concomitant with the rise in TPN-associated apoptosis, a significant decrease in the mrna expression of the antiapoptotic factor Bcl-2. Additionally, a decline in the proapoptotic factors Bak and Bad was noted. No significant change in the Bax expression was seen. It has been reported in many tissues, that alterations in the ratio of Bcl-2 to Bax can determine the

KGF DECREASES TPN-INDUCED INTESTINAL APOPTOSIS 95 development of apoptosis. 16 It seems that, more important than the level of expression of individual anti- or proapoptotic proteins, it is the relative proportion of death agonists or antagonists, and this ratio determines how a cell will respond to an apoptotic signal. 17 Our finding of a decrease in Bcl-2 caused a shift in the Bcl-2 to Bax ratio from 1.00 to 0.17 in control and TPN mice, respectively. The pro- or antiapoptotic function of these proteins may also depend on the cell type, the apoptotic stimuli, the cellular context (eg, cell cycle dependence of the process), and the cellular environment (eg, presence or absence of growth factors). 15 Because we observed a significant increase in EC apoptosis, the decrease of the proapoptotic members Bak and Bad, which itself would decrease apoptosis, may be subsided by either the distinctive decrease of the antiapoptotic member Bcl-2, or other determinants contributing to apoptosis. We have shown in a previous study, that IEL s FasL expression is significantly increased with TPN. These results suggest that the extrinsic pathway may play an important role in apoptosis with TPN administration. 1 Future studies will be necessary to better understand the exact mechanism of TPN-induced apoptosis. Little is known of the antiapoptotic mechanism of KGF. A recent study on alveolar cells reported that KGF treatment in vitro led to a decreased expression of Bax, p21, and p53; however, Bcl-2 expression was not affected. 8 Another report showed that in KGF-treated keratinocytes the Fas/FasL pathway is still highly prone to be activated if triggered with apoptotic stimuli, suggesting that KGF action does not operate via the extrinsic pathway. 18 KGF led to a significant increase in the mrna expression of the antiapoptotic factors Bcl-2 and Bcl-w. Because of the marked increase in mrna expression and its critical role in directing apoptosis, Bcl-2 protein expression was confirmed with Western blotting. This showed a 2.6-fold increase compared with TPN alone. We showed in our study, that the mechanism by which KGF prevented EC apoptosis includes the upregulation of the antiapoptotic member Bcl-2. Clearly, further investigation is needed to better define the mechanism of KGF and its special effect on intestinal EC apoptosis. TPN-induced apoptosis led to a decrease in Bcl-2 mrna expression. KGF administration to TPN mice led to a decrease in EC apoptosis and an increase in the expression of the anti-apoptotic factor Bcl-2. Modalities to increase secretion of KGF by IEL, or exogenous KGF administration, may be beneficial in patients on TPN. ACKNOWLEDGMENT Recombinant human KGF was a gift of Amgen Corporation. 1. Yang H, Fan Y, Teitelbaum DH: Interferon- gamma mediates epithelial cell apoptosis through a Fas/FasL interaction in a mouse model of total parenteral nutrition (TPN). Gastroenterology 120:A676, 2001 2. Ciccocioppo R, Di Sabatino A, Gasbarrini G, et al: Apoptosis and gastrointestinal tract [Review]. Italian Journal of Gastroenterology & Hepatology 31:162-172, 1999 3. Boismenu R, Havran WL: Modulation of epithelial cell growth by intraepithelial gamma delta T cells. Science 266:1253-1255, 1994 4. Antony PA, Yang H, Fan Y, et al: Altered expression of intraepithelial lymphocytes (IEL) keratinocyte growth (KGF) mrna in the mouse. Gastroenterology 118:685, 2000 5. Estivariz CF, Jonal CR, Gu LH, et al: Gut-trophic effects of keratinocyte growth factor in rat small intestine and colon during enteral refeeding. JPEN J Parenter Enter Nutri 22:259-267, 1998 6. Kiristioglu I, Teitelbaum DH: Alteration of the intestinal intraepithelial lymphocytes during total parenteral nutrition. J Surg Res 79:91-96, 1998 7. Li J, Gocinski B, Henken B, et al: Effects of parenteral nutrition on gut-associated lymphoid tissue. J Traum 39:44-52, 1995 8. Buckley S, Barsky L, Driscoll B, et al: Apoptosis and DNA damage in type 2 alveolar epithelial cells cultured from hyperoxic rats. Am J Physiol 274:L714-20, 1998 9. Hall PA, Coates PJ, Ansari B, et al: Regulation of cell number in the mammalian gastrointestinal tract: The importance of apoptosis. J Cell Sci 107:3569-3577, 1994 REFERENCES 10. Klein RM, McKenzie JC: The role of cell renewal in the ontogeny of the intestine. J Pediatr Gastroenterol Nutr 2:204-228, 1983 11. Iwakiri D PD: Keratinocyte growth factor promotes goblet cell differentiation through regulation of goblet cell silencer inhibitor. Gastroenterology 120:1372-1380, 2001 12. Goodlad RA, Mandir N, Meeran K, et al: Does the response of the intestinal epithelium to keratinocyte growth factor vary according to the method of administration? Regulatory Peptides 87:83-90, 2000 13. Senaldi G, Shaklee CL, Simon B, et al: Keratinocyte growth factor protects murine hepatocytes from tumor necrosis factor-induced apoptosis in vivo and in vitro. Hepatology 27:1584-1591, 1998 14. Stennicke HR, Salvesen GS: Caspases Controlling intracellular signals by protease zymogen activation. Biochim Biophys Acta 1477:299-306, 2000 15. Jones BA, Gores GJ: Physiology and pathophysiology of apoptosis in epithelial cells of the liver, pancreas, and intestine. Am J Physiol 273:G1174-1188, 1997 16. Ina KI, Fukushima K, Kusugami K, et al: Resistance of Crohn s disease T cells to multiple apoptotic signals is associated with a bcl-2/bax mucosal imbalance. J Immunol 163:1081-1090, 1999 17. Kroemer G: The proto-oncogene Bcl-2 and its role in regulating apoptosis. Nat Med 3:614-620, 1997 18. Freiberg R, Spencer D, Choate K, et al: Specific triggering of the Fas signal transduction pathway in normal human keratinocytes. J Biol Chem 271:1666-1669, 1996

96 WILDHABER, YANG, AND TEITELBAUM Discussion P. Tam (Hong Kong): I would like to congratulate you on a beautiful paper setting up a difficult set of experiments. Did you look to see if there was any difference across the small bowel between the jejunem or the ileum in your results? Are there some areas that respond better to your stimulant factor. B.E. Wildhaber (response): We did not study separate parts of the small bowel in this report, and all my samples are from the jejunum. We felt that the maximum effect of KGF would be on the jejunum, and, thus, for this initial study felt that this would be the best area of bowel to study. It would be very interesting to investigate this in future studies.