Neonatal necrotizing enterocolitis (NEC), one of the major causes of morbidity and mortality in premature infants, usually occurs days or even weeks after birth, when premature infants receive enteral feeding and bacterial colonization of the colon is established (1,2). Once colonization is established, in the anaerobic environment of the colon, unabsorbed carbohydrates, such as lactose, can be rapidly fermented into gases (hydrogen, carbon dioxide, and methane) and short-chain fatty acids (SCFAs), mainly acetic acid, butyric acid, and propionic acid (3,4). The absorption of SCFAs coupled with sodium absorption is one of the major mechanisms for salt and water uptake in the colon (5). SCFAs also serve as a major energy source for colonic epithelial cells and exert a significant trophic effect on colonic mucosa (6,7). In premature infants, production of SCFAs begins when enteral feeding is introduced. Because premature infants cannot digest a significant proportion of the lactose in the formula because of relative lactase deficiency in the small intestine, bacterial fermentation in the colon becomes a significant pathway for lactose digestion (8). Up to one third of the lactose taken by healthy premature infants may be fermented into SCFAs by bacteria and then absorbed (9). In addition to SCFAs, other organic acids such as lactic acid are produced by specific lactic acid–producing bacterial species. These bacteria have been recognized to have many beneficial effects on the gastrointestinal tract of human infants and have been used as probiotics (10).
During normal conditions, organic acids are rapidly absorbed by the colon and play important roles in intestinal biology. However, in premature infants, an abnormal state of organic acid production may arise as a result of carbohydrate malabsorption and/or bacterial overgrowth (11). Production of organic acids may exceed the buffering and absorptive capacity of the colon, leading to an increased concentration of organic acids in the colon. Organic acids in the distal ileum may also increase because of reflux through the ileocecal valve and/or local bacterial overgrowth (12,13). A high concentration of organic acids can significantly lower the pH of the intestinal lumen contents. The high concentration of organic acids along with the low pH in the intestine may cause intestinal mucosal injury. In fact, acetic acid in high concentration can produce intestinal mucosal injury in animals (14,15). We have therefore speculated that organic acid–induced intestinal mucosal injury may be important in the pathogenesis of NEC. The aim of this study was to examine and compare the effects of the major colonic organic acids—acetic acid, butyric acid, and lactic acid—on the intestinal mucosa of newborn rats when these acids were administered per rectum.
MATERIALS AND METHODS
Newborn Sprague-Dawley rats (Taconic Laboratory, NY) were housed with their mothers and maintained in alternate 12-hour periods of darkness and light.
At the age of 10 days, the rat pups were weighed and randomly divided into 7 experimental groups. Littermates were usually assigned to different experimental groups. A total of 72 animals from 8 litters was used. A 3.5F umbilical catheter was inserted per rectum 4.0 cm deep into the proximal colon (the total length of the colon in 10-day-old rat pups is approximately 4.0 cm). Acetic acid, butyric acid, and lactic acid (Sigma, St. Louis, MO) at concentrations of 150 mmol/L and 300 mmol/L were used. The pH of all solutions was adjusted to 4.0 with NaOH and the relative volume administered by rectal tube was the same for each experimental solution—.1 ml/10g body weight. A similar volume of normal saline, with pH also adjusted to 4.0 with HCL, was administered as a control. Therefore, a total of seven groups were created: a normal saline control group (n = 10), 150 mM acetic acid group (n = 11), 300 mM acetic acid group (n = 11), 150 mM butyric acid group (n = 10), 300 mM butyric acid group (n = 11), 150 mM lactic acid group (n = 7), and 300 mM lactic acid group (n = 12). The administration of the organic acids was not blinded to the investigators because of their obvious odor. In each individual experiment, various organic acid solutions were administered into different groups on the same day. Animals were housed with and cared for by their mothers before and after the experiments.
Twenty-four hours after the luminal administration of the organic acid solutions or normal saline, the rats were weighed and killed by cervical dislocation. The proximal 3 cm of colon was removed and weighed. A portion of the proximal colonic tissue and the adjacent distal ileum was fixed in 4% neutral phosphate-buffered formalin, cross-sectioned at 2-mm intervals, and embedded in paraffin. A 5-μm-thick section from each block was stained with hematoxylin and eosin. Two pathologists who were blinded to the solutions received by each rat evaluated all the slides and gave histologic injury scores to all the intestinal tissues. The scores ranged from 0 to 5 based on the degree of injury: 0 for normal histology; 1 for minimal to mild inflammation without necrosis; 2 for moderate to severe inflammation without necrosis; 3 for mucosal necrosis and/or distorted/regenerative architecture indicative of previous severe epithelial injury; 4 for mucosal and submucosal necrosis; and 5 for transmural necrosis.
The research protocol was approved by the Institutional Animal Care and Use Committee. The data are presented as mean values and SEM. Analysis of variance was used for statistical analysis. The difference between each group was compared with the Tukey test or Kruskal-Wallis test when the data were not normally distributed. P < 0.05 was considered significant.
The baseline body weights were the same in all groups: 20.1 ± 0.6 g, 19.6 ± 0.6 g, 20.7 ± 0.5 g, 19.3 ± 0.5 g, 21.2 ± 0.7 g, 19.8 ± 0.7 g, and 20.6 ± 0.7 g for the normal saline, 150 mM acetic acid, 300 mM acetic acid, 150 mM butyric acid, 300 mM butyric acid, 150 mM lactic acid, and 300 mM lactic acid groups, respectively (P > 0.05). All rats survived for 24 hours after luminal administration of all of the agents. The percent body weight changes among all groups at 24 hours are presented in Figure 1. Normal saline–treated control rats gained a significant amount of body weight (14.4 ± 0.9%). Rats receiving either 150 or 300 mM lactic acid gained similar body weight. However, rats receiving 300 mM acetic or butyric acid had significant weight loss compared with the normal saline controls (−1.5 ± 2.3% and −2.1 ± 2.2%, respectively;P < 0.05). Rats receiving 150 mM acetic acid or 150 mM butyric acid gained less body weight, but the changes were not statistically significant from the normal saline control group.
At the time they were killed, the rats that had received 300 mM acetic acid or butyric acid had similar grossly evident intestinal edema and patchy hemorrhage, especially in the proximal colon and distal ileum (Fig. 2). The proximal 3-cm colon wet weights in those two groups were similar to each other and significantly heavier than the control group (P < 0.05). The colon wet weights in the other groups of rats that received 150 mM acetic acid, 150 mM butyric acid, or 150mM and 300 mM lactic acid were comparable to the control group (Fig. 3).
In the majority of animals receiving 300 mM or 300 mM butyric acid, histologic examination revealed intestinal necrosis of some degree, with the depth of involvement in most cases extending down to the submucosa or muscularis propria (histologic scores of 3–5). The areas of necrosis were associated with dense neutrophilic infiltrates. Of the animals receiving 150 mM acetic acid or 150 mM butyric acid, only 4 of the 21 showed intestinal necrosis, which was confined to the mucosa and accompanied by acute inflammation. All remaining samples in those groups were either histologically normal or showed only mild acute inflammation, consisting of slight neutrophilic infiltrates within the lamina propria. There was no appreciable difference between the injury induced by acetic acid and butyric acid. No histologic injuries were identified in the ileum or colon in the rats that received either 150 mM or 300 mM lactic acid at either 150 mmol/L or 300 mmol/L. All intestinal tissues from the normal saline control group were normal. The mean histologic injury scores for distal ileum and colon from all seven groups are presented in Figures 4 and 5, respectively. Representative photomicrographs from different histologic injury scores are presented in Figure 6.
Prematurity, formula feeding, and bacterial colonization have been recognized as major risk factors for NEC (2). Clark et al. (16) suggested that intraluminal biochemical changes caused by bacterial fermentation of unabsorbed carbohydrates may be responsible for the development of NEC in premature infants. By examining the intestinal contents at or adjacent to the site of perforation or necrosis in 17 infants requiring surgery for advanced NEC, they found 16 of 17 patients with NEC had intraluminal contents with a pH of less than 5 (pH range, 3.8–4.6). However, organic acid concentrations were not measured. Increased breath hydrogen excretion is found in patients with NEC even before the onset of clinical symptoms (17,18), supporting the theory that bacterial fermentation products may promote the development of NEC in premature infants. Pneumatosis intestinalis in NEC patients is also thought to be secondary to hydrogen gas produced by bacterial fermentation (19).
In our study, we have demonstrated that luminal administration of acetic acid or butyric acid, two of the main SCFAs produced by bacterial fermentation of undigested carbohydrates, produces dose-dependent intestinal mucosal injury in newborn rats at pH 4.0, a pH chosen to resemble the acidic milieu encountered in NEC (16). Similar levels of injury are demonstrated for both acetic acid and butyric acid, with the injury involving the colon and the distal ileum and resembling human NEC histologically and clinically (20). Based on histology, the injury appeared to be acute inflammation and necrosis. Furthermore, additional experiments performed later showed that the colonic tissue myeloperoxidase activity was significantly increased in the rats with butyric acid–induced intestinal mucosal injury compared with that of normal saline–treated control animals (data not shown). The tissue inflammation and edema, decreased oral intake and increased tissue breakdown may contribute to the increase in colon wet weight and the decrease in daily body weight gain in the rats that received high-dose SCFAs. The colon weight and the change in daily body weight gain might possibly be simple and objective parameters to quantify the degree of intestinal injury in the experimental situation. Relative immaturity in the gastrointestinal tract of newborn rats and the completion of gut bacterial colonization in 10 days' postnatal age may make our study relevant to NEC in premature infants.
However, whether endogenous luminal SCFAs can produce a similar injurious effect in premature infants requires further study. A beneficial effect of the presence of SCFAs in the colon has been speculated (21). SCFAs have even been proposed as therapy for patients with inflammatory bowel disease and experimental colitis (22–25). Paradoxically, butyrate can also lead to apoptosis and is cytotoxic in high concentrations (26,27). The concentration of luminal SCFAs in some premature infants may be increased because of overproduction to a level high enough to cause intestinal mucosal injury. Inability to clear the intraluminal SCFAs because of poor intestinal motility in premature infants may also contribute to the pathology of NEC. This would explain all three major risk factors for NEC, i.e., prematurity, enteral feeding and bacterial colonization. Unfortunately, no data are available on the physiologic concentration of SCFAs in the proximal colon and distal ileum of premature infants receiving enteral feeding. Furthermore, we do not know the colonic luminal SCFA concentration in those premature infants who demonstrated increased bacterial fermentation before the development of NEC (17,18). The total SCFA concentration in the stool of normal full-term infants is most likely less than 100 mmol/L (28). However, the concentration of SCFAs in proximal colon content is usually much higher than that of the distal colon (29–31). The concentration of total SCFAs in proximal colon in premature infants may reach 200 mmol/L or even 300 mmol/L when maldigestion and poor motility are present. Since both acetic acid and butyric acid have similar injurious effects, the range of total SCFA concentration and prolonged long-term exposure may cause intestinal mucosal injury, leading to NEC in premature infant.
Luminal administration of lactic acid, another major organic acid produced in the colon via bacterial fermentation of undigested carbohydrates, at the same concentration and pH, did not cause gross or microscopic intestinal injury in our model. The mechanism for this difference in effect is not clear. Our findings are different from those of Argenzio et al. (32). They found that both sodium acetate and sodium lactate in concentration of 100 mmol/L at pH 4.0 induce similar reversible injury to porcine colon. In their study, both salt solutions but not the organic acid solutions were used, and the pH of the solutions was adjusted with H3PO4, an inorganic acid. The difference may also be a result of different methods or animal species used in the experiments. Furthermore, after pH adjustment with NaOH, the final osmolarity for both butyric acid 300 mmol/L and acetic acid 300 mmol/L solutions was about 340 mOsm/L. This osmolarity alone was unlikely to produce the mucosal injury seen in our experiments. For reassurance, additional experiments were performed. We found that in the normal saline control or 300 mM lactic acid group, increasing the final osmolarity to 500 mOsm/L with sodium chloride or sodium lactate did not produce any identifiable mucosal injury (data not shown).
Our findings may be correlated with the clinical observation that premature infants on breast milk feeding have a much lower incidence of NEC (33). In fact, breast milk promotes lactic acid–producing bacterial colonization so that breast-fed babies produce more lactic acid, while formula fed infants have higher concentrations of SCFAs in their stools (28,34). Administration of lactic acid–producing bacterial species can increase the production of lactic acid and reduce the concentration of SCFAs (35). Some of the probiotics have been shown to reduce the incidence of NEC in experimental animal models as well as in one clinical trial (36–38). We speculate that the beneficial effect of probiotics may be a result of preferential production of lactic acid instead of SCFAs by the probiotics. Further studies are needed to validate this speculation. A better understanding of the luminal production of lactic acid and SCFAs by intestinal flora in premature infants, and their role in the pathogenesis of NEC, may lead to the development of novel strategies to prevent NEC.
The authors thank Professor Rami Eliakim and Dr. Robert Green for the advice and careful review of the manuscript.
1. Caplan MS, Jilling T. New concepts in necrotizing enterocolitis
. Curr Opin Pediatr 2001; 13:111–115.
2. Wilson R, Kanto WP, McCarthy BJ, Feldman RA. Age at onset of necrotizing enterocolitis
: Risk factors in small infants. Am J Dis Child 1982; 136:814–816.
3. Bond JH, Levitt MD. Fate of soluble carbohydrate in the colon of rats and man. J Clin Invest 1976; 57:1158–1164.
4. Kien CL. Colonic fermentation of carbohydrate in the premature infant: Possible relevance to necrotizing enterocolitis
. J Pediatr 1990; 117:S52–58.
5. Gordon HA. The role of the intestinal flora in absorption: A comparative study between germ-free and conventional animals. In: Csaky TZ, ed. Intestinal absorption and malabsorption. New York: Raven Press, 1975, pp 238–251.
6. Roediger WEW. Utilization of nutrients by isolated epithelial cells of rat colon. Gastroenterology 1982; 83:424–429.
7. Sakata T. Stimulatory effect of short-chain fatty acids on epithelial cell proliferation in the rat intestine: A possible explanation for trophic effects of fermentable fiber, gut microbes and luminal trophic factors. Br J Nutr 1987; 58:95–103.
8. MacLean Jr, WC Fink BB. Lactose malabsorption by premature infants: Magnitude and clinical significance. J Pediatr 1980; 97:383–388.
9. Kien CL, McClead RE, Cordero Jr. L In vivo lactose digestion in premature infants. Am J Clin Nutr 1996; 64:700–705.
10. Vanderhoof JA. Probiotics and intestinal inflammatory disorders in infants and children. J Pediatr Gastroenterol Nutr 2000; 30:S34–S38.
11. Kien CL. Digestion, absorption, and fermentation of carbohydrates in the newborn. Clin Perinatol 1996; 23:211–228.
12. Hoverstad T, Bjorneklett A, Fausa O, Midtvedt T. Short-chain fatty acids in the small-bowel bacterial overgrowth syndrome. Scand J Gastroenterol 1985; 20:492–499.
13. Lloyd DR, Brown JD, Brown GA, Booth IW. Elevated short chain fatty acid concentrations in anaerobic small bowel contamination. Acta Paediatr 1992; 81:51–56.
14. MacPherson BR, Pfeiffer CJ. Experimental production of diffuse colitis in rats. Digestion 1978; 17:135–150.
15. Kim HS, Berstad A. Experimental colitis in animal models. Scand J Gastroenterol 1992; 27:529–537.
16. Clark DA, Thompson JE, Weiner LB, McMillan JA, Scneider AJ, Rokahr JE. Necrotizing enterocolitis
: Intraluminal biochemistry in human neonates and a rabbit model. Pediatr Res 1985; 19:919–921.
17. Garstin WIH, Boston VE. Sequential assay of expired breath hydrogen as a means of predicting necrotizing enterocolitis
in susceptible infants. J Pediatr Surg 1987; 22:208–210.
18. Cheu HW, Brown DR, Rowe M. Breath hydrogen excretion as a screening test for early diagnosis of necrotizing
enterolitis. Am J Dis Child 1989; 143:156–159.
19. Engel RR, Virning NL, Hunt CE, Levitt MD. Origin of mural gas in necrotizing enterocolitis
[abstract]. Pediatr Res 1973; 7:292.
20. Ballance WA, Dahms BB, Shenker N, Kliegman RM. Pathology of neonatal necrotizing enterocolitis
: A ten-year experience. J Pediatr 1990; 117:S6–13.
21. Mortensen PB, Clausen MR. Short-chain fatty acids in the human colon: Relation to gastrointestinal health and disease. Scand J Gastroenterol Suppl 1996; 216:132–148.
22. Patz J, Jacobsohn WZ, Gottschalk-Sabag S, Zeides S, Braverman DZ. Treatment of refractory distal ulcerative colitis with short chain fatty acid enemas. Am J Gastroenterol 1996; 91:731–734.
23. Scheppach W, German-Austrian SCFA Study Group. Treatment of distal ulcerative colitis with short-chain fatty acid enemas: A placebo-controlled trial. Dig Dis Sci 1996; 41:2254–2259.
24. Scheppach W, Christl SU, Bartram HP, Richter F, Kasper H. Effects of short-chain fatty acids on the inflamed colonic mucosa. Scand J Gastroenterol Suppl 1997; 222:53–57.
25. Segain JP, Raingeard de la Bletiere D, Bourreille A, et al. Butyrate inhibits inflammatory responses through NFκB inhibition: Implications for Crohn's disease. Gut 2000; 47:397–403.
26. Avivi-Green C, Polak-Charcon S, Madar Z, Schwartz B. Apoptosis cascade proteins are regulated in vivo by high intracolonic butyrate concentration: Correlation with colon cancer inhibition. Oncol Res 2000; 12:83–95.
27. Popoff MR, Jolivet-Reynaud C, Carlier JP. Cytotoxic activity of Clostridium butyricum
supernatants induced by butyrate. FEMS Microbiol Lett 1987; 43:95–100.
28. Ogawa K, Ben RA, Pons S, de Paolo MIL, Fernandez LB. Volatile fatty acids, lactic acid
and pH in stools of breast-fed and bottle-fed infants. J Pediatr Gastroenterol Nutr 1992; 15:248–252.
29. Cummings JH, Pomare EW, Branch WJ, Naylor CPE, MacFarlane GT. Short chain fatty acids
in human large intestine, portal hepatic and venous blood. Gut 1987; 28:1221–1227.
30. Lupton JR, Kurtz PP. Relationship of colonic luminal short chain fatty acids
and pH to in vivo cell proliferation in rats. J Nutr 1993; 123:1522–1530.
31. Holtug K, Rasmussen HS, Mortensen PB. An in vivo study of short-chain fatty acid concentrations, production and absorption in pig (Sus scrofa) colon. Comp Biochem Physiol Comp Physiol 1992; 103:189–197.
32. Argenzio RA, Meuten DJ. Short-chain fatty acids induce reversible injury of porcine colon. Dig Dis Sci 1991; 36:1495–1468.
33. Neu J. Necrotizing enterocolitis
: The search for a unifying pathogenic theory leading to prevention. Pediatr Clin North Am 1996; 43:409–432.
34. Edwards CA, Parrett AM, Balmer SE, Wharton BA. Faecal short chain fatty acids
in breast-fed and formula-fed babies. Acta Paediatr 1994; 83:459–462.
35. Bianchi-Salvadori B, Vesely R, Ferrari A, Canzi E, Casiraghi C, Brighenti F. Behaviour of the pharmacetical probiotic preparation VSL#3 in human ileostomy effluent containing its own natural elements. New Microbiol 2001; 24:23–33.
36. Butel MJ, Roland N, Hibert A, et al. Clostridial pathogenicity in experimental necrotising enterocolitis
in gnotobiotic quails and protective role of bifidobacterial. J Med Microbiol 1998; 47:391–399.
37. Caplan MS, Miller-Catchpole R, Kaup S, et al. Bifidobacterial supplementation reduces the incidence of necrotizing enterocolitis
in a neonatal rat model. Gastroenterology 1999; 117:577–583.
38. Hoyos AB. Reduced incidence of necrotizing enterocolitis
associated with enteral administration of Lactobacillus acidophilus
and Bifidobacterium infantis
to neonates in an intensive care unit. Int J Infect Dis 1999; 3:197–202.