Fetal intestinal transplant as an accessory enteral segment

Pediatr Surg Int (1997) 12:367-369

© Springer-Verlag 1997

B. H . Giiven~ • T. S a l m a n • B. Tokar • E. Siirmen T. Altu~ " A. (yelik

Fetal intestinal transplant as an accessory enteral segment

Accepted: 6 June 1996 Abstract Fetal tissue transplantation has gathered considerable interest among researchers dealing with organ transplantation. A large number of studies concerning fetal intestinal transplantation have been published in the past 2 decades, almost all of them aiming to determine the feasibility of a properly functioning fetal transplant in continuity with the host's own enteral system. This study was designed to determine the absorptive capacity of the neogut in vivo, without anastomosing the transplant to the host's intestine, and to evaluate its use as an accessory enteral segment. Intestinal segments taken from Wistar albino fetuses were transplanted subcutaneously into the abdominal wall of 20 Sprague-Dawley rats. Immunosuppression was maintained by daily cyclosporin A (Cy A) 10 mg/kg injections s.c. and evaluated by determination of serum Cy A level and T-helper/T-suppressor cell ratio. The neogut was converted into a Thiry-Vella loop 2 weeks after transplantation. A test solution composed of 20% glucose and Trophamine was perfused via the stomas; glucose and amino acid absorption gradients were calculated. The gamma-glutamyl transferase (GGT) activity and mitotic index of the neogut were determined. Results were compared to those obtained from the host. There was no significant difference (P > 0.05) in glucose absorption between the neogut and the host tissue. Amino acid absorption and specific G G T activity were

Presented at the 39th Annual International Congress of the British Association of Paediatric Surgeons, Leeds B. H. Gfiven~ ([]) Selami Cesme Guzel Sok. 16/14, TR-81030 Kiziltoprak, Istanbul, Turkey B. H. Gfiveng. T. Salman-B. Tokar-E. Sfirmen. T. Altu~' A. ~elik Department of Pediatric Surgery, University of Istanbul, Istanbul School of Medicine, Istanbul, Turkey

significantly less (P < 0.01) in the neogut. There was no significant difference (P > 0.05) between neogut and host intestine in mitotic index. Our data support the idea of using a transplanted fetal intestinal segment as an accessory feeding route. Key words Fetal tissue transplantation • Small intestinal

transplantation

Introduction The energy requirements of a growing child are best met by a properly functioning intestinal system. It has been established that the loss of more than 80% of the small intestine is incompatible with life. Intravenous administration of vital nutrients means survival for the patient suffering from short-bowel syndrome (SBS). During this supplementary treatment, the intestine is expected to adapt to its new physiological state. Parenteral nutrition, on the other hand, is associated with serious complications such as thrombosis, sepsis, and liver impairment in addition to psychological side effects and high cost [3, 15, 16]. Intestinal transplantation, on the other hand, has presented even more complicated problems. Numerous efforts have shown that clinical success is elusive while problems such as rejection, graft-versus-host disease, and sepsis due to immunosupression await solution [1, 8, 14]. The search for alternative treatments has led researchers to investigate the use of fetal tissues in transplantation. Like many other investigators who have responded to Zinzar et al.'s original paper [20], we have shown in previous studies that an avascular fetal intestinal segment may survive and continue to grow with an intact lumen in the subcutaneous space of a syngeneic host [13]. Deutsch et al. [2] and Kellnar et al. [5] succeeded in anastomosing the neogut to the host's own enteral system. Studies accomplished so far have shown that animals with such transplants are unable to eat

368

solid food; problems of motility have been made responsible for failure of the model [17]. This study was designed to determine the absorptive capacity of the neogut in vivo in an allogeneic host under immunosuppressive therapy and to evaluate its use as an accessory enteral segment without anastomosing it to the host's enteral system.

fluorescence polarization immunoassay with a TDx Analyzer (Abbott Diagnostic Products). Blood samples were taken from the heart prior to death. The serum T-helper/T-suppressor cell (T4/TS) ratio was determined using MCA48F mouse monoclonal antibody to rat T-suppressor cytotoxic cells-CD8 and MCA55F mouse monoclonal antibody to rat T-helper and macrophages-CD4 kits (Serotec). Cells were counted by a fluorescence-activated cell sorting device. All data were expressed as mean + standard error (SEM). Statistical analyses were done using the Mann Whitney U test; P < 0.05 was considered significant.

Materials and methods Under ether inhalation anesthesia, Wistar albino fetuses (1517 days of gestation) were delivered by cesarean section. Under tenfold magnification, approximately 4-cm jejunoileal segments of 20 fetuses were harvested. The mesentery was carefully stripped and the lumen was flushed with cold saline solution. The segment was then placed heterotopically in the subcutaneous space of the abdomen of 20 Sprague-Dawley (120-150 g) rats through a lower transverse incision. Each animal received cyclosporin A (Cy A) 10 mg/kg subcutaneously daily. Two weeks following the initial operation, with the animals under anesthesia, the matured segments were converted into Thiry-Vella loops using interrupted 5/0 silk sutures. Daily intraluminal saline irrigations were started immediately afterward. Carbohydrate and amino acid absorption were evaluated using a closed perfusion technique by an indirect method. The test solution was composed of 12.5 ml 20% glucose, 45 ml Trophamine, 0.5 ml 20% saline sermn, and 2 ml KC1, and was designed to resemble the daily caloric requirement of a rat. On the 8th day following the enterocutaneostomy, 20 ml of the solution was perfused at a rate of 0.159 ml/min, through the Thiry-Vella loop continuously for 2 h using a Harvard pump. An identical jejunoileal segment was then prepared from the host's own intestine as a control, taking care to preserve the mesenteric vessels. The test solution was likewise perfused through the ends of the isolated loop. The glucose and amino acid concentrations of the solution were calculated after each perfusion. The ratio of the difference in concentrations to the concentration of the perfusate was determined as the indirect absorbtion gradient for glucose and amino acids. The following formula was used for absorbtion gradient calculation: Amino acid or glucose concentration before perfusion concentration

Amino acid or glucose concentration after perfusion

Results All of the transplanted segments revascularized and were viable at the end of 2 weeks. They continued to grow as longitudinal tubes with an intact lumen until the study was terminated. During the study, 3 rats died due to complications from anesthetics and 4 due to immunosuppressive therapy and were replaced. Each transplanted segment bled during the enterocutaneostomy demonstrating an effective blood supply from the neovascularized zone. The mucoid secretions changed in character following daily irrigations, gaining a clear appearance. Glucose absorption of the neogut and host intestine were 12.04 + 5.48% and 18.75 + 10.56%, respectively. Amino acid absorption was 7.15 + 4.62% for the neogut and 25.05 + 13.42% for the host intestine. Amino acids were absorbed significantly better (P < 0.001) from the host's own intestine, while glucose absorption showed no statistically significant difference (Fig. 1). The enzyme activities of the host and fetal gut were 33.973 + 10.002 and 14.731 + 4.144 IU/g protein, respectively; enzyme

45-

x 100

35-

Amino acid or glucose concentration before perfusion Glucose concentration was calculated using Trinder's [18] glucose oxidase method, and amino acid concentration using Khachadurian's [6] amino acid nitrogen detern?ination method. The activity of gamma-glutamyl transpeptidase (GGT), an enzyme responsible for transportation of amino acids through the intestinal mucosa, was analyzed according to Rosalki et al.'s [12] method. In 10 rats the mucosa of the transplant and host intestine was freed from the muscular coat. The protein content of the homogenate was studied according to Lowry et al.'s method [11]. The enzymatic activity was expressed in IU micromolar p-nitroanilide liberated per min per g protein. The results were compared with the host's own intestinal enzymatic activity. The reproductive capacity of the crypt cells was determined using the vincristine-induced metaphase arrest technique [4]. Each rat was given 1 mg/kg vincristine sulfate (Oncovin) intraperitoneally 30 min prior to deathl Specimens of intestine taken from transplant and host tissue were fixed in 10% formalin and embedded in paraffin blocks. Sections were stained with hematoxylin-eosin and examined under a light microscope. Mitotic activity was expressed as the number of crypt cells arrested during metaphase per 100 cells counted in a single area [4]. In order to evaluate the effectiveness of the immunosupressive therapy, serum Cy A was calculated 8-12 h following injection by

[ ] Transplant

4030

25" 200

(n 15" .Q

< 10" 5i

Amino acid

Glucose

Fig. 1 Amino acid and glucose absorption gradients of the perfusate (mean + SEM) Table 1 Brush-border enzyme activity and mitotic activity in fetal

rat intestine in an allogeneic host (mean + SEM)

Neogut Host

Transpeptidase activity (IU/protein)

Mitotic activity (n/100 cells)

14.731 + 4.144 33.973 +_ 10.002 p < 0.01

4.6 + 1.42 4.8 + 1.t0 p > 0.05

369 activity was significantly higher in the host tissue (P < 0.01) (Table 1). The productive capacity of the neogut was successfully preserved. Mitotic indices in the host tissue and neogut were 4.6 + 1.42 and 4.8 + 1.10 cells in a 100-cell population, respectively. There was no significant difference between the two groups (Table 1). The microscopic studies showed no significant cellular infiltration, only a few p o l y m o r p h o n u c l e a r cells (PMN). The mean serum Cy A level 315 + 90.3 ng/ml, was within the suggested optimal range of 200-400 ng/ml. The T4/T8 cytotoxic ratio was 1.65 + 0.26, which also indicates that there was no rejection.

Discussion The numerous experiments on intestinal transplantation accomplished so far have mostly concentrated on immunological problems such as the rejection process [9, 19]. The functional integrity of a transplant, on the other hand, is just as i m p o r t a n t as immunosuppression. There is controversy as to how problems like lymphatic drainage, reinnervation, and absorptive defects of the intestine should be solved. In the meantime, the patient has to be protected f r o m the side effects of parenteral nutrition, which m a y lead to hepatic failure. In our model, we did not intend to anastomose the neogut to the host's enteral system. Our aim was to determine whether it could be used as an accessory enteral segment for possible nutritional substitution. We observed that the absorptive gradient of glucose in the transplanted segment did not differ significantly f r o m that of the host, which shows that glucose was absorbed through the m u c o s a o f the neogut. The significant difference in amino acid absorption, on the other hand, m a y have been due to the absence of numerous factors such as the enzymes produced in the d u o d e n u m or a lack of sufficient substrate stimulation. The difference in G G T activity between the neogut and the host intestine roughly correlated with the results obtained for amino acid absorption. Our measurements of serum Cy A showed optimal serum levels, within the margins given in previously performed studies [7, 10]. The T4/T8 cytotoxic ratios correlated closely with serum Cy A levels, and additionally supported the adequacy of the i m m u n o s u p p ressive therapy. Microscopic studies of the transplants showed that the mucosa was composed of mixed epithelial cells, submucosa, and a bilayered muscular coat similar to normal ileum. There was only minimal cellular infitration, which consisted mostly of P M N . The mitotic index was similar c o m p a r e d to host tissue. All o f these results demonstrated a vital transplant. The mitotic index and T4/T8 cytotoxicity ratio m a y be useful parameters in monitoring transplanted intestinal segments. In conclusion, our data support the idea of using a fetal intestinal segment as an accessory feeding system;

further studies, however, are needed to refine the functional enhancement of the neogut with regard to absorption and brush-border enzyme studies. The addition of missing duodenal secretions and brush-border enzymes to the substrate m a y increase the absorbtive capacity of the neogut.

References 1. Abu-Elmagd K, Todo S, Tzakis A, Reyes J, Nour B, Furukawa H, Fung JJ, Demetris A, Starzl TE (1994) Three years clinical experience with intestinal transplantation. J Am Coll Surg 179: 385-400 2. Deutsch AA, Leapman SB, Arensman K, et al (1973) Anastomosis of fetal rat intestine to the normal intestine of the host. J Surg Res 15:176-181 3. Hale DA, Waldorf KA, Kleinschmidt J, et al (1991) Small intestinal transplantation in nonhuman primates. J Pediatr Surg 26:914-920 4. Holt PR, Yeh Kwo-Yih (1988) Colonic proliferation is increased in senescent rats. Gastroenterology 95:1556-1563 5. Kellnar S, Herkomer C, Bae S, et al (1990) Allogeneic transplantation of fetal rat intestine: anastomosis to the normal bowel of the host. J Pediatr Surg 25:415-417 6. Khachadurian A, Knox WK, Cullen AM (1960) Colorimetric ninhydrin method for total alpha-amino acid of urine. J Lab Clin Med 56:321-332 7. Kimura K, Larosa CA, Blank MA, et al (1990) Successful segmental intestinal transplantation in enterectomized pigs. Ann Surg 211:158-164 8. Kirkman RL (1984) Small bowel transplantation. Transplantation 37:429-433 9. Kobayashi E, Toyama N, Kiyozaki H, Enosawa S, Walker N, Miyata M (1994) Small bowel transplantation for pediatric short bowel syndrome: evaluation of the graft length required for development and the immunologic aspects relating to graft length. J Pediatr Surg 29:1331-1334 10. Lee KKW, Schraut WH (1986) Structure and function of orthotopic small bowel allografts in rats treated with cyclosporinc. Am J Surg 151:55-60 11. Lowry OH, Rosenbrough NJ, Fau AL (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265269 12. Rosalki SB, Rau D, Lehmann D (1970) Determination of serum gamma-glutamyl transpeptidase activity and its clinical applications. Ann Biol Chem 7:143-148 13. Salman T, Given9 BH, Celik A, et al (1990) Fetal intestinal transplantation in rats. Turk J Med Biol Res 1:167-169 14. Schraut WH (1988) Current status of small bowel transplantation. Gastroenterology 94:525-538 15. Schwartz MZ, Maeda K (1985) Short bowel syndrome in infants and children. Pediatr Clin North Am 32:1265-1279 16. Takagi Y, Okada A (1994) Candidates for small bowel transplantation: our experience and a survey of home parenteral nutrition in Japan. Transplant Proc 26:1446 17. Tisinai K, Shedd F, Harris R, et al (1990) Comparison of growth, neovascularization and enzymatic function of fetal intestinal grafts in the omentum and renal capsule. J Pediatr Surg 25:914-916 18. Trinder P (1969) Determination of blood glucose using an oxidase peroxidase system. Ann Clin Biochem 6:24-28 19. Yang R, Liu Q, Rescoria FJ, Grosfeld JL (1994) Lack of graftversus-host disease after fetal intestine transplantation. J Pediatr Surg 29:1157-1160 20. Zinzar SN, Leitina BI, Tumyan BG, et al (1971) Very large organ-like structures formed by syngeneic fetal alimentary tract transplanted as a whole or in parts. Rev Eur Etudes Clin et Biol 16:455-458