4.

4.1.

 4. .............................................1 4.1. .............................................................................................................................2 4.2. . .......................................................................................................4 4.2.1. ......................................................................................................................5 4.2.1.1. ......................................................................................6 4.2.1.2. . ...................................................................................29 4.2.1.3. . ....................................................................................34 4.2.1.4. ...................................................................34 4.2.2. . .............................................................................................................35 4.2.2.1. ..............................36 4.2.2.2. . .................................................................................................................36 4.2.2.3. . ............................................................................................................36 4.2.2.4. . ................................................................................................................36 4.2.2.5. . ..................................................................................................................36 4.2.2.6. (). .....................................37 4.2.2.7. ...........................................................37 4.2.3. . ....................................................................................................................38 4.2.3.1. . ................................................................39 4.2.3.2. . ..................................................................39 4.2.3.3. . ........................................................................................................39 4.2.3.4. .........................................................................................................39 4.2.3.5. . ....39 4.2.3.6. . ............................................................................................39 4.2.3.7. . .........................................................39 4.3. . .....................................................................................40

4.2. .

4.2.1.

4.2.1.1. . Lactulose causes a decrease in blood ammonia concentration and reduces the degree of portalsystemic encephalopathy. These actions are considered to be results of the following: Bacterial degradation of Lactulose in the colon acidifies the colonic contents. This acidification of colonic contents results in the retention of ammonia in the colon as the ammonium ion. Since the colonic contents are then more acid than the blood, ammonia can be expected to migrate from the blood into the colon to form the ammonium ion. The acid colonic contents convert NH3 to the ammonium ion (NH4)+, trapping it and preventing its absorption. The laxative action of the metabolites of Lactulose then expels the trapped ammonium ion from the colon. Experimental data indicate that Lactulose is poorly absorbed. Lactulose given orally to man and experimental animals resulted in only small amounts reaching the blood. Urinary excretion has been determined to be 3% or less and is essentially complete within 24 hours. When incubated with extracts of human small intestinal mucosa, Lactulose was not hydrolyzed during a 24-hour period and did not inhibit the activity of these extracts on lactose. Lactulose reaches the colon essentially unchanged. There it is metabolized by bacteria with the formation of low molecular weight acids that acidify the colon contents.

Effect of lactulose on establishment of a rat non-alcoholic steatohepatitis model. / Fan JG, Xu ZJ, Wang GL. // World J Gastroenterol. 2005 Aug 28;11(32):5053-6. Department of Gastroenterology, Shanghai First People's Hospital, Jiaotong University, Shanghai 200080, China. fanjg@citiz.net AIM: To explore the relationship between changes of intestinal environment and pathogenesis of non-alcoholic steatohepatitis (NASH). METHODS: Forty-two Sprague-Dawley rats were randomly divided into model group (n = 24), treatment group (n = 12), and control group (n = 6). The rats of model and treatment groups were given high-fat diet, and those of the control group were given normal diet. Furthermore, the rats of treatment group were given lactulose after 8 wk of high-fat diet. Twelve rats of the model group were killed at 8 wk of high-fat diet. At the 16 wk the rats of treatment group, control group, and the rest of the model group were killed. The serum levels of aminotransferase were measured and the histology of livers was observed by H and E staining. RESULTS: The livers of rats presented the pathological features of steatohepatitis with higher serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in the model group after 16 wk. Compared to the model group, the serum levels of ALT and AST in

treatment group decreased significantly and were close to the normal group, and the hepatic inflammation scores also decreased markedly than those in the model group after 16 wk (5.83+/2.02 vs 3.63+/-0.64, P<0.05), but were still higher than those in the model group after 8 wk (3.63+/-0.64 vs 1.98+/-0.90, P<0.05). However, the degree of hepatic steatosis had no changes in treatment group compared to the model group after 16 wk. CONCLUSION: Lactulose could ameliorate the hepatic inflammation of rats with steatohepatitis induced by fat-rich diet, but could not completely prevent the development of steatohepatitis. It is suggested that intestinal environmental changes such as intestinal bacteria overgrowth, are one of the important factors in the pathogenesis of NASH. INTRODUCTION The pathogenesis of non-alcoholic steatohepatitis (NASH) remains unclear[1-5]. Several studies have suggested that small intestine bacterial overgrowth might play a role in NASH[6-9]. NASH is a common complication of jejuno-ileal bypass for morbid obesity[10-12]. NASH has also been described in adults during total parenteral nutrition (TPN) and in multiple jejunal diverticulae with bacterial overgrowth in small intestine[13,14]. However, evidence is insufficient to indicate that intestinal flora has much to do with the usually insidious process of NASH[15-18]. Recently, it was reported that the prevalence of small intestine bacterial overgrowth is high in obese patients with NASH, as assessed by the 14C-D-xylose-lactulose breath test[15], suggesting that intestinal environmental changes such as small intestine bacterial overgrowth and gut original endotoxemia may play an important role in the development of NASH. To further study the relationship between the change of intestinal environment and pathogenesis of NASH, we investigated the effect of lactulose on the establishment of a rat NASH model. MATERIALS AND METHODS Animals Male Sprague-Dawley rats (n = 42) weighing 140-160 g were purchased from the Shanghai Experimental Animal Center (Shanghai, China), and housed in cages under standard conditions with free access to water. After being fed with standard rat chow for 1 wk, the animals were randomly divided into control, model, and treatment groups. The control group (n = 6) received normal diet, the model group (n = 24) and treatment group (n = 12) received fat-rich diet (normal diet plus 2% cholesterol and 10% lard). Eight weeks later, the treatment group was administrated a solution of lactulose syrup (Solvay Pharma, China) instead of water. In general, the dose of lactulose syrup for adults was 45 mL/d. If the average weight of adults was 59 kg, the dose of lactulose liquid was 0.9 mL/(kgd). According to the dose of adults, rats might be given a 10-fold higher dose (3.7 mL/d). Livers and blood samples were collected from 12 rats of model group at the wk 8. At wk 16, samples were collected from other rats in three groups. In brief: after fasting and water deprivation for more than 12 h, the rats were weighed and anesthetized with 1% pentobarbital by intraperitoneal injection. Then the blood samples were collected through abdominal aorta, the liver tissues were weighed and fixed in 40 g/L formaldehyde, embedded in paraffin. Biochemical measurement

The serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), triglycerides (TG), and total cholesterol (TC) were measured using an automatic biochemical analytical system. Serum endotoxin level Serum endotoxin level was measured by chromogenic limulus amoebocyte lysate test in Shanghai Clinical Test Center, and the tubes without pyrogen were supplied by the center. Hepatic histology The liver tissue sections were stained with hematoxylin and eosin. Each section was assessed under 1020 light microscopic fields and scored for the severity of steatosis and inflammation according to the following criteria: Steatosis was scored as: grade 0: no fat; grade 1: fatty hepatocytes occupying less than 33% of the hepatic parenchyma; grade 2: fatty hepatocytes occupying 34-66% of the hepatic parenchyma; and grade 3: fatty hepatocytes occupying more than 66% of the hepatic parenchyma. The diagnosis of fatty liver could be confirmed, when fatty hepatocytes occupying more than 33% of the hepatic parenchyma. Portal inflammation (P), intralobular inflammation (L), piecemeal necrosis (PN), and bridging hepatic necrosis (BN) had a score from 1 to 4 according to the pathologic severity. PN and BN had a greater correlation with the prognosis, the score was two times higher than P and L. The inflammation score was P+L+2PN+2BN. Statistical analysis All results were expressed as meanSD. Statistical differences between means were determined by Students t test. Rank sum test was used in enumeration of data. P<0.05 was considered statistically significant. RESULTS During the experiment, no rat died in three groups. In the treatment group, there were no marked increase in the volume of feces. The feces were soft. No significant difference was found in the body weights between model group and control group. But the ratio of the liver wet weight with the body weight in the model group increased significantly than that in the control group. In the treatment group, the body weights were lower than those in the model group at wk 16, and the ratio of the liver wet weight with the body weight (liver exponent, %) significantly decreased compared to that in the model group at wk 8 and 16 (Table 1).

Serum lipid In model group, the serum level of TC was markedly higher than that in control group, but the serum level of TG was similar to that in control group. No significant difference in serum levels

of TC between the model group and the treatment group, but the serum level of TG in the treatment group decreased markedly (Table 2).

Serum aminotransferase The serum levels of ALT and AST in the model group had an increasing tendency at wk 8, but no significant difference was found between the model group and control group until wk 16. The serum levels of ALT and AST in treatment group decreased significantly, and almost became normal (Table 3).

Serum endotoxin level The serum endotoxin level in portal vein was higher than that in abdominal aorta, both in model group and control group (P<0.05), but there was no significant difference in serum endotoxin level both in portal vein and in abdominal aorta between the two groups. Therefore, we did not measure the serum endotoxin level in treatment group (Table 4).

Hepatic histology The livers of control group had normal morphology, while the livers of model group and treatment group became yellow, dull, enlarged, fragile, and full. Microscopically, the livers of control group had no marked abnormality. At wk 8, the livers of model group were engorged with microvesicular and macrovesicular fat. Fatty liver could be diagnosed in 11 of 12 rats. Four out of twelve rats had mild intralobular inflammation. At wk 16, hepatic steatosis was much severe, and intralobular areas were infiltrated by mixed inflammatory cells in model group. These lesions were located mainly in zone 3 areas. Intralobular inflammation was more severe than portal inflammation. Some livers had several large areas of necrosis melted by focal intralobular inflammation. Two rats had hepatic piecemeal necrosis and three had bridging necrosis. In treatment group, the degree of hepatic steatosis slightly decreased, while the score of hepatic inflammation activity significantly decreased compared to those in model group. No

hepatic piecemeal necrosis and bridging necrosis were found in treatment group (Tables 5 and 6).

DISCUSSION The histological characteristics of NASH resemble those of alcoholic steatohepatitis, suggesting that both diseases may have a similar pathogenesis and can benefit from similar therapies[1518]. Studies in alcohol-fed rodents have shown that intestinal bacteria, bacterial endotoxin, and endotoxininducible cytokine can modulate alcohol-induced liver damages including hepatic necrosis and fibrosis[19,20]. Treatment with antibiotics and lactobacillus that inhibit production of endotoxin by the intestinal flora can significantly inhibit the development of steatohepatitis in alcohol-fed animals[21,22]. Although the intestinal flora is known to play a critical role in the pathogenesis of alcohol-related liver disease, its role in NASH has been poorly understood. Surgical procedures (such as jejunoileal bypass) and TPN cause intestinal stasis and secondary bacterial overgrowth, accelerate the progression of fatty liver disease in obesity patients, suggesting that increased exposure to intestinal bacterial products may contribute to the pathogenesis of NASH[1-5]. It has been reported that the prevalence of small intestinal bacterial overgrowth is high in patients with NASH[15]. Lactulose could be fermented by colonic bacteria and turn into lactic acid and acetic acid, which can lower colonic pH, diminish ammonia production and normalize intestinal transit; therefore, lactulose syrup can be used in treatment of hepatic encephalopathy. Lactulose promotes the growth of acidophilous lactobacilli, bifidobacteria, and Grampositive bacteria, while inhibits Gram-negative bacteria and prevents gut-derived endotoxemia[23]. In order to explore the relationship between the change of intestinal environmental and pathogenesis of NASH, we observed the effect of lactulose on NASH rats. The treatment group was given lactulose after 8 wk of high-fat diet, when simple fatty liver developed in rats. The model group developed NASH after 16 wk, while treatment with lactulose for 8 wk improved both serum aminotransferase and hepatic inflammation. These results suggest that intestinal environmental changes, such as small intestinal bacterial overgrowth, increased intestinal permeability and subsequent gut-derived endotoxemia may play an important role in the development of NASH. Although the treatment group received lactulose for 8 wk, the hepatic inflammation scores were still higher than those in model group, indicating that lactulose can ameliorate hepatic inflammation, but cannot prevent NASH. Lactulose could not improve

hepatic steatosis in treatment group, suggesting that the change of intestinal environment is not closely related to hepatic steatosis. In our serial researches, we found that the serum endotoxin level in NASH rats were significantly elevated, when hepatic fibrosis occurred after 24 wk of high-fat diet. The serum endotoxin level in portal vein and peripheral vessels had no significant difference between model group and control group, but the serum endotoxin level in portal vein was markedly higher than that in peripheral vessels in both groups. These results suggest that SD rats might have mild endotoxemia. We also found that the expression of endotoxin receptorsCD14 and toll-like receptor 4 was upregulated in model group, suggesting that Kupffer cell sensitivity to endotoxin increases and low doses of endotoxin might injure liver. Oral administration of lactulose may improve hepatic inflammation of NASH rats by reducing serum endotoxin level in portal vein. Furthermore, other bacterial products such as peptidoglycanpolysaccharide polymers rather than endotoxin, could stimulate Kupffer cells and injure liver, because bacterial species rather than aerobic Gram-negative bacteria such as Escherichia coli may play a role in the pathogenesis of small intestinal bacterial overgrowth[23,24].

Comparison of probiotics and lactulose in the treatment of minimal hepatic encephalopathy in rats. / Jia L, Zhang MH. // World J Gastroenterol. 2005 Feb 14;11(6):90811. Department of Digestive Diseases, Affiliated First People's Municipal Hospital, Guangzhou Medical College, Guangdong, China. gzjialin@hotmail.com AIM: To compare the efficacy of probiotic preparation Golden Bifid and lactulose on rat experimental model of minimal hepatic encephalopathy (MHE) induced by thioactamide (TAA). METHODS: MHE was induced by intraperitoneal injection of TAA (200 mg/kg) every 24 h for two consecutive days. Thirty-six male MHE models were then randomly divided into 3 groups: TAA group (n = 12) received tap water ad libitum only; lactulose group (n = 12) and probiotics group (n = 12) were gavaged, respectively with 8 mL/kg of lactulose and 1.5 g/kg of probiotic preparation Golden Bifid (highly concentrated combination of probiotic) dissolved in 2 mL of normal saline, once a day for 8 d (from the 5(th) d before the experiment to the 3(rd) d of the experiment). The latency of brainstem auditory evoked potentials (BAEP) I was used as an objective index of MHE. The incidence of MHE, the level of serum endotoxin, ammonia, liver function and histological grade of hepatic injury of rats were examined individually. RESULTS: There were no overt HE and rat deaths in 3 groups. The incidence of MHE, the levels of blood ammonia and endotoxin in TAA group, which were 83.3% (10/12), 168.33+/15.44 mg/dL and 0.36+/-0.04 EU/mL, respectively, were significantly higher than those in lactulose group, which were 33.3% (4/12), 110.25+/-7.39 mg/dL and 0.19+/-0.02 EU/mL, and probiotics group, which were 33.3% (4/12), 108.58+/-10.24 mg/dL and 0.13+/-0.03 EU/mL respectively (P<0.001). It showed that either probiotics or lactulose could significantly lower the level of hyperammonemia and hyper-endotoxemia, lighten centrolobular necrotic areas as well as inflammatory reaction in the liver of rats, normalize the latency of BAEP, and decrease the incidence of MHE. However, no significant differences were observed between these two groups (P>0.05).

CONCLUSION: Probiotic compound Golden Bifid is at least as useful as lactulose for the prevention and treatment of MHE. Probiotic therapy may be a safe, natural, well-tolerated therapy appropriate for the long-term treatment of MHE.

Effect of lactulose on short-chain fatty acids and lactate production and on the growth of faecal flora, with special reference to Clostridium difficile. / Ito Y, Moriwaki H, Muto Y, Kato N, Watanabe K, Ueno K. // J Med Microbiol. 1997 Jan;46(1):80-4. First Department of Internal Medicine, Gifu University School of Medicine, Japan. Lactulose exerts a beneficial effect on hepatic encephalopathy by decreasing toxic short-chain (iC4-nC6) fatty acid (isobutyrate, butyrate, isovalerate, valerate, isocaproate and caproate) production. However, the precise mechanism by which lactulose exerts this effect remains uncertain. This study investigated the effect of lactulose on faecal flora, particularly Clostridium difficile, which produces mostly iC4-nC6 fatty acids. An in-vitro faecal incubation system was used to estimate how lactulose influences production of short-chain (C2-nC6) fatty acids and lactate. Faecal specimens were collected from patients with liver cirrhosis, who carried C. difficile in the colon. Supplementation of lactulose along with blood in faecal specimens decreased iC4-nC6 fatty acids production and increased acetate and lactate production, resulting in increased faecal acidity. These changes were statistically significant when compared with supplementation by blood alone. Quantitative faecal culture demonstrated that lactulose supplementation suppressed the growth of C. difficile and Bacteroides spp. (B. fragilis group), iC4-nC6 fatty acids-producing organisms. These results suggest that decreased faecal levels of iC4-nC6 fatty acids after lactulose supplementation may be related to suppression of iC4-nC6 fatty acids-producing faecal organisms, especially C. difficile. Introduction Cerebrotoxic substances derived from the gut such as ammonia, short-chain fatty acids (SCFAs), mercaptans and y-aminobutyric acids are thought to be responsible for hepatic encephalopathy (HE). While ammonia is most widely studied among these toxic substances, hyperammonaemia is not always associated with HE. Increased blood levels of isobutyrate (iC4), butyrate (nC4), isovalerate (iC5), valerate (nC5), isocaproate (iC6) and caproate (nC6) in patients with HE [l] and induction of HE by these SCFAs in animals [2] have been reported. Lactulose is a disaccharide (/3 1-4 galacto-fructose) which has been used extensively since 1966 in the treatment of HE [3]. Taken orally, lactulose passes unchanged into the colon where it is hydrolysed by bacterial action to organic acids, principally acetate and lactate. Although the precise mode of action of lactulose remains unclear, proposed mechanisms are: (i) lowering colonic pH, thereby decreasing the production of ammonia by bacteria [4] and the absorption of non-ionised ammonia [5]; (ii) serving as substrate to increase the incorporation of ammonia into bacterial protein [6]; and (iii) decreasing the intestinal transit time available for production and absorption of ammonia because of its cathartic effect [7]. Mortensen et al. [8] showed that lactulose decreased the degradation of amino acids, albumin and blood to iC4-nC6 fatty acids by its acidifying effect.

The majority of SCFAs are produced by anaerobic bacteria that ferment dietary carbohydrates to SCFAs in the colon [9]. These bacteria include Clostridium spp., Bacteroides spp. (B. fragilis group), Fusobacterium spp. and Megasphaera spp. Among them, the Bacteroides spp. are one of the major components of intestinal flora, contributing to ammonia production but producing only a small amount of iC5. In contrast, C. perfringens produces a large amount of nC4 and C. dfjcile produces large amounts of iC4-nC6 fatty acids [lo, 111.] A previous study showed an overgrowth of C. d@cile and C. perfringens in the faecal flora of patients with HE, which was induced by administration of antimicrobial agents [12]. The rate of carriage of C. dficile was c. 20% in patients with liver cirrhosis (unpublished observations) and 7% in healthy elderly adults [ 131. Although these observations indicate that Clostridium spp. are of clinical importance in the development of HE, the effect of lactulose on C. perfringens has been examined in only a few studies [14, 151 and no report has focused on C. dzjicile. The aim of this study was to evaluate the effect of lactulose on the faecal levels of SCFAs and lactate as well as on the growth of intestinal flora, especially C. difJiZe, with an in-vitro faecal incubation system, where faecal specimens containing C. d@cile were emp 1 oyed. Materials and methods Patients Six inpatients with liver cirrhosis, who carried C. difficile, provided stool specimens. No antimicrobial agents had been prescribed during the 4 weeks prior to the study. Clinical characteristics and laboratory data on admission are shown in Table 1. Two patients had mild diarrhoea while others had normal faeces. Preparation of samples The in-vitro faecal incubation system was similar to that of Mortensen et aZ. [16]. Freshly passed faeces were homogenised with five volumes of 100 mM NaCl and 50mM KC1. Ten ml of faecal homogenates were mixed with one of the following; blood (0.5 mll10 ml faecal homogenate), lactulose (25 mmol/L) and blood (0.5 mV10 ml faecal homogenate) or no addition. Each faecal sample was then incubated anaerobically at 37C for 24 h in an atmosphere of N2 80%, H2 10% and c02 10%.

Bacteriological examination One ml of the incubated faecal sample was homogenised with 9 ml of anaerobic diluent, and serial 10-fold dilutions were made to give a final dilution of lo7. Each odd number of logarithmic dilution was inoculated quantitatively onto the media shown in Table 2. Aerobic and anaerobic cultures were incubated at 37C for 72 h. The number of organisms was expressed as loglo cfdml of incubated faecal sample. Isolates were identified to genus level according to standard methods [lo, 11, 171. Determination of SCFAs and lactate in incubated faecal samples One ml of the incubated faecal sample was mixed with 0.1 ml of H2SO4 50% and 1 ml of diethyl ether, and centrifuged briefly to make an ether extract for the determination of SCFAs - acetate (C2), propionate (C3), iC4, nC4, iC5, nC5, iC6 and nC6 [lo]. For the determination of lactate, a mixture of 1 ml of the incubated faecal sample, 0.1 ml of H2S04 50% and 1 ml of (CH30H)2BF3 was heated at 100C for 5 min make a chloroform extract. These extracts were subjected to gas-liquid chromatography (Shimadzu GC-l4A, Shimadzu Corp., Kyoto, Japan) with a 1.6 m X 3.2-mm column packed with Reoplex 400 10% on Chromosorb W(AWPMCS) (80- 100 mesh). Gas flows were 50 ml/min for N2, 50 ml/min for H2 and 1000 ml/min for air. Temperatures were kept at 150C for the column and 220C for the injector and the flame-ionisation detector. Production of SCFAs and lactate by isolates from faeces A standard loopful of isolate was inoculated into Gifu Anaerobic Medium (GAM) broth [ 181 (Nissui Pharmaceutical Co., Ltd) and incubated anaerobically at 37C for 3-4 days. SCFAs and lactate production were determined as described above. pH ana lys is The pH of incubated faecal samples was measured with pH test paper (Advantec@, Toyo Roshi Co. Ltd, Tokyo, Japan). Statistical methods All comparisons were performed by paired t test with Statview software (Abacus Concepts Inc., Berkeley, CA, USA). Results SCFAs and lactate levels in incubated faecal s a mp 1 es Following incubation, the faecal samples of all six individuals showed production of C2 -nC6 fatty acids. The addition of blood increased acetate, iC4-nC6 and lactate production while addition of lactulose extinguished the effect of blood on iC4-nC6 fatty acids production and increased faecal levels of acetate and lactate (Table 3). A marked decrease in faecal pH was observed following the addition of lactulose. Changes in faecal flora The growth of C. dfficile, Bacteroides spp., BIJidobacterium spp. and Enterobacteriaceae was significantly suppressed by the addition of lactulose, while growth of C. perfringens was not suppressed (Table 4).

Lactobacillus spp. became the major component afterlactulose was added to faecal samples. SCFAs and lactate production by faecal isolates Bacteroides isolates produced iC5, Fusobacterium and C. pe$+zgens isolates produced nC4, while C. difJicile isolates produced iC4-nC6 fatty acids. Discussion Although the precise mode of action of lactulose has not been clarified, one of the speculated mechanisms is the acidification of the colon contents by fermented products of lactulose. Mortensen et al. [16] showed that iC4-nC6 fatty acid production by faecal flora was increased by the addition of blood, albumin and amino acids but decreased by adding lactulose in vitro. They initially suggested that this effect was the result of substrate competition, i.e., colonic bacteria preferred lactulose to blood when both were present. Later it was reported that the acidifying effect of lactulose was more important in decreasing iC4-nC6 fatty acids production [8]. The present study confirmed that the addition of lactulose decreased faecal levels of iC4-nC6 fatty acids, increased acetate production and reduced faecal pH. It further demonstrated that lactulose caused an increase in lactate production in the faecal incubation system, which contributed to an increase in faecal acidity. Of particular importance was the finding that the

growth of both C. dfjcile (mostly iC4-nC6 fatty acid-producing) and Bacteroides spp. (iC5producing) was suppressed by lactulose. The growth of C. dffcile is particularly susceptible to changes in colonic acidity [19]. Rolfe et al. [20] showed that, in vitro, Lactobacillus spp. and group D enterococci produced large amounts of lactate that inhibited the growth of C. diffile. They also suggested that SCFAs played an important role in the induction of colonisation resistance against C. dficile in a conventional hamster model [21]. The invitro study by May et al. [22] suggested that acidification with SCFAs produced by fermentation of dietary fibre suppressed C. dfjcile. The study data can be interpreted as follows: acidification with acetate and lactate produced by fermentation of lactulose suppressed faecal anaerobes, especially C. difJicile, resulting in a decrease of iC4- nC6 fatty acids production. Others have reported the effect of oral lactulose on human faecal flora in vivo. An increase in Lactobacillus and Bijidobacterium spp. and a decrease in Bacteroides spp. and coliform organisms (Enterobacteriaceae) were noted [ 15,231. These in-vivo findings are similar to the invitro results of the present study except for Bijidobacterium spp. The conflicting BiJidobacterium spp. data seem to result from a difference in environmental pH. The pH in the right colon decreases to 4.85 after ingestion of lactulose [24], whereas the pH was below 4 in the in-vitro faecal incubation system. As reported by Vince et al. [6], when the pH was uncontrolled and acidic conditions developed, Bzjidobacteriurn spp. were suppressed by the addition of lactulose to samples. In any event, the action of lactulose in increasing the number of acid (lactate and/or acetate)-producing bacteria such as L. acidophilus, Enterococcus spp. or Bifidobacterium spp. seems to be important. Administration of these bacteria was shown to have an effect on HE [25-271. Furthermore, a poorly absorbed disaccharide, lactitol, which was shown to improve HE, induces colonic acidification to a similar degree to lactulose [28]. Oligosaccharides taken orally significantly increased the number of BiJidobacterium spp. in human faecal flora [29]. Dietary fibre, which escapes from host digestion and is fermented to acetate, propionate and butyrate in the colon [9], was shown to increase the number of acid-producing bacteria but decrease the number of C. dificile and resultant toxin [22]. A vegetable protein diet supplemented with dietary fibre was also effective in HE [30]. In addition to lactulose, these substances may also exert a beneficial effect on HE by suppressing the growth of the iC4-nC6 fatty acids-producing organisms, C. dzfJicile and Bacteroides spp., through colonic acidification. Further invivo studies are needed to establish the mechanisms of action of these substances.

Effect of protein and lactulose on the production of gamma-aminobutyric acid by faecal Escherichia coli. / al Mardini H, al Jumaili B, Record CO, Burke D. // Gut. 1991 Sep;32(9):1007-10. Gastroenterology Unit, Royal Victoria Infirmary, Newcastle Upon Tyne.

The value of lactulose treatment in hepatic encephalopathy is widely recognised but its mode of action remains controversial. Much evidence supports a role for gamma-aminobutyric acid in hepatic encephalopathy, and lactulose could alter its bacterial production in the gut. Using the rat synaptic membrane assay and gas chromatography mass spectrometry, the production of gammaaminobutyric acid by faecal Escherichia coli, with and without the addition of albumin, haemoglobin, whole blood, and lactulose under aerobic and anaerobic conditions was determined. Using an inorganic medium, maximal gamma-aminobutyric acid production occurred after a culture period of between 25 and 50 hours. The concentration after 30 hours of aerobic culture at 37 degrees C by a single strain was mean (SEM), 101 (5) mumol/l (99% confidence intervals 87-114 mumol/l; n = 8; interassay coefficient of variation 14.7%). gammaaminobutyric acid production was significantly increased by the addition of albumin and haemoglobin. Under anerobic conditions, it was one fifth of that produced aerobically, but the addition of albumin and haemoglobin increased production by greater than 700%. Lactulose did not significantly attenuate gamma-aminobutyric acid production under aerobic or anaerobic conditions. gamma-aminobutyric acid determined by the rat synaptic membrane assay showed a highly significant correlation (r = 0.99) with that detected by gas chromatography mass spectrometry. These data confirm that gamma-aminobutyric acid is produced by faecal E coli and that protein enhances its production considerably, and suggest that lactulose does not exert its therapeutic effect by attenuating gamma-aminobutyric acid production. Cerebrotoxic substances derived from the gut are thought to be involved in the mediation of hepatic encephalopathy. Treatments shown to benefit this disorder are thought to act by an effect on colonic bacterial metabolism. Although lactulose has been used since 1966' in the treatment of hepatic encephalopathy, its mode of Gastroenterology Unit, action remains controversial. Proposed mechanisms include acidification of the colon,2 decrease in ammonia production, 3 normalisation of the fatty acid profile,4 or merely as a purgative. y-aminobutyric acid is the predominant neuroinhibitory neurotransmitter in the mammalian brain and recent interest has highlighted it as a candidate mediator of hepatic encephalopathy. Raised values of y-aminobutyric acid like upon activity have been found in the plasma of patients with liver disease67 and in experimental animals.' y-aminobutyric acid arising in the gut is thought to cross the blood brain barrier because of the defective blood brain permeability that has previously been shown in a rabbit galactosamine model of hepatic encephalopathy.9 Schafer et al' showed that faecal bacteria are capable of producing y-aminobutyric acid like substances in vitro and suggested that the colonic flora were the source of the y-aminobutyric acid like activity detected in the blood. Other workers" have suggested that the bowel mucosa itself contributes to y-aminobutyric acid like activity detected in the portal blood. Gastrointestinal bleeding and high protein diets are potent precipitations of hepatic encephalopathy and it is thought that the resulting modulation of colonic bacterial metabolism may lead to the production of cerebrotoxic substances. The aims of this study were to examine the production of yaminobutyric acid by faecal Escherichia coli, determine the effect of protein on its production, and to assess whether lactulose exerts its beneficial effect by attenuating yaminobutyric acid production. Methods

Faecal E coli were isolated from normal controls and stored on nutrient agar slopes. Ten strains of the isolated E coli were cultured separately for 18 hours at 37C in tryptic soya broth in order to determine the amount of y-aminobutyric acid like activity produced by each strain. The strain that produced the highest amount was chosen for further study. To determine the most suitable growth medium, the selected strain of E coli was grown under aerobic conditions at 37C for 18 hours in different liquid culture media, brain heart infusion broth, proteose peptone No 3 (Difco), casamino acid yeast extract medium, tryptone soya broth, and a synthetic non-organic medium (NaCl 5*5 g/l; KCI 3-7 g/l; CaCl2 0 15 g/l; MgCl2 0-1 g/l; NH4SO4 1 5 g/l; K2PO4 0 27 g/l; glucose 10-0 g/l; Tris buffer 12*1 g/l; pH 7 4). Incubation of e coli with whole blood, albumin, haemoglobin, and lactulose The effect of adding protein to the culture media was studied by incubating E coli alone and together with whole blood (O 5 ml/10 ml), haemoglobin (60 mg/10 ml), or albumin (120 mg/ 10 ml). Lactulose was added to the culture medium alone at concentrations of 25, 50, and 100 mmol/l, and together with albumin (120 mg/ 10 ml) at a concentration of 100 mmol/l (all substrates were added to the culture medium immediately before incubation). Media with and without the various additives were incubated in the absence ofE coli as controls. E coli were grown both aerobically in a shaking water bath at 37C and in an anaerobic cabinet at 37C. At the end of the culture period viable counts were determined by the method of Miles and Misra. 12 The remaining culture medium was centrifuged at 3000 g at 4C and aliquots of supernatant stored at -20C pending analysis of pH and concentrations of y-aminobutyric acid, glucose, and lactulose. The lactulose concentration was determined by gas liquid chromatography by the method of Laker.'3 Y-aminobutyric acid analysis y-aminobutyric acid like activity was measured by a radioreceptor assay using rat synapatic membranes and radiolabelled y-aminobutyric acid as described by Enna et al'4 and Ferkany et al.'5 Samples were assayed in triplicate and residual radioactivity assessed using a 1B liquid scintillation counter. The concentration of true y-aminobutyric acid was also determined by gas chromatography mass spectrophotometry using a Finnigan Mat 1020 based on the method described by Moroni et al'6 and Levy et al."7 A DB5 (30 mx2 5 mm internal diameter) gas chromatography column was used with an initial column temperature of 80C for two minutes. The final temperature of 260C for five minutes was achieved by increasing the temperature by 8C per minute. Statistical analysis The results of y-aminobutyric acid obtained by the two techniques were expressed as mean (SEM) and the statistical differences between the groups were assessed using unpaired Student's t test (two tailed). Linear regression analysis was used to determine the correlation between yaminobutyric acid concentrations measured by the radioreceptor assay and by gas chromatography mass spectrophotometry. Results

Effect of different culture media and length of incubation on the production of y-aminobutyric acid like activity Table I shows the background yaminobutyric acid like activity found in the culture media used. Because all four standard media contained detectable amounts of y-aminobutyric acid like activity, the non-organic medium, which showed no background activity, was chosen for use in future studies. The mean y-aminobutyric acid like activity found after 30 hours aerobic E coli culture in this medium at 37C was 101 (5) iimol/l (n=8). The interassay coefficient of variation was 14-7%. yaminobutyric acid like activity was detectable in small amounts after two to 12 hours incubation (Fig 1), while there was an exponential increase from 1-2 ,umol/l at 12 hours to 100 ,umol/l at 25 hours. Thereafter, there was only a slight increase until the end of the incubation period (50 hours). Based upon these observations, a 30 hour incubation period at 37C was used in the following studies. Production of y-aminobutyric acid like activity under aerobic conditions The different concentrations of lactulose (25, 50, and 100 mmol/1) led to sequential increases in the amount of y-aminobutyric acid like activity produced from 0-27 (n=2) to 0-37 (n=2) and 0-43 mol/bacteriumx 10"'5 (n=2) respectively. Lactulose at a concentration of 100 mmol/l was used in further studies. Figure 2 shows the concentration of yaminobutyric acid like activity ([tmol/l) under aerobic and anaerobic conditions after the addition of protein and lactulose. Under aerobic conditions there was a significant increase from mean (SEM), 104 (11) (E coli alone) to 145 (10) and 286 (4) iimol/l after culture in media containing blood (p<0 05) and haemoglobin (p<O0OOl) respectively, and 327 (8) iimol/l when albumin was added to the culture medium (p<0001). There was no significant increase with the addition of lactulose alone (134 (9 3) timol/l, p=NS), but after the addition of lactulose and albumin to the medium the amount produced (272 (11-9) iimol/l) was significantly less than that produced by albumin alone (p<O01).

When the results were expressed as the amount of y-aminobutyric acid like activity produced per bacterium (Table II), there was an increase from 0-26 (0 01) mol/bacteriumx 10-l' by E coli alone to 0 37 (0 03) with the addition of blood (p<0 05), 0 39 (0 06) with lactulose (p<0Q05), 1-65 (0 27) with haemoglobin (p<0001), and 0-88 (0 08) mol with albumin (p<O 001). However, the addition of lactulose to albumin (086 (0 16) mol/bacterium x 10-'5) did not seem to alter significantly the amount of y-aminobutyric acid like activity produced per bacterium in comparison with albumin alone. Production of y-aminobutyric acid like activity under anaerobic conditions The concentrations of y-aminobutyric acid like activity found after incubation under anaerobic conditions, with and without the addition of lactulose and protein, are shown in Figure 2. In general, y-aminobutyric acid like activity was less than that obtained under aerobic conditions. There was a similar increase in activity from mean (SEM) 23 (2) iimolll by E coli alone to 77 (3) ,umol/l after the addition of blood (p<0 001), to 186 (13) with albumin (p<0001) and 214 (20) ,umol/l with haemoglobin (p<0001). Again the addition of lactulose alone (28 (4) [tmol/1) showed no significant effect and the addition of lactulose and albumin (179 (16) [tmol/1) was not significantly different from that found with albumin alone.

When the results were expressed as yaminobutyric acid equivalent per bacterium (Table II) there were noticeable differences compared with the values obtained under aerobic conditions. The production of yaminobutyric acid like activity (molx 10-'5) was 9 7 (1-9) byE coli alone, 14 9 (6 9) with lactulose (NS), 45-5 (8 3) with albumin (p<0 01), 46A4 (6&9) with albumin and lactulose (p<0 01), 132.8 (28'9) with haemoglobin (p<0 01) and 3-2 (0-6) with the addition of blood. Again, the difference between the values obtained from media containing albumin alone and albumin and lactulose were not significant. y-aminobutyric acid concentrations measured by radioreceptor assay compared with gas chromatography mass spectrophotometry The results obtained by both the radioreceptor assay and gas chromatography mass spectrophotometry are shown in Figure 3. There was excellent correlation between the values obtained by the two techniques (r=0 99). Effect of incubation on lactulose concentration

The concentration of lactulose after incubation of E coli with lactulose alone under aerobic and anaerobic conditions did not change significantly from the starting concentration. In the presence of albumin and lactulose, however, the concentrations after 30 hours' incubation in duplicate cultures were 47 and 53 mmol/l under aerobic conditions and 40 and 40 mmol/l under anaerobic conditions, indicating lactulose utilisation in the presence of protein. Discussion Lactulose has long been used in the management of patients with hepatic encephalopathy,' and more recently the related compound lactitol (which is more palatable) has also been shown to be effective in patients with both overt and subclinical encephalopathy. 18 19 Lactulose is known to be metabolised by colonic bacteria20 and its metabolism results in the reduction of colonic pH and alteration in colonic volatile fatty acids.2' These do not, however, explain directly its beneficial effect in the management of this condition. In our hands, the four different liquid culture media used in previous studies'" gave high background counts of y-aminobutyric acid, making them unsuitable for studying changes in activity during culture. The non-organic media used gave minimal background activity, although compared with Minuk's findings,22 the production of y-aminobutyric acid by E coli was delayed, reaching its maximal stationary value after 30 hours of incubation. The addition of blood, haemoglobin, albumin, and lactulose to the culture media increased the production of y-aminobutyric acid by E coli under aerobic conditions. Under anaerobic conditions, although less yaminobutyric acid was produced, its production increased by more than 700% after the addition of albumin and haemoglobin. When the amount of y-aminobutyric acid produced was related to the luxurience of bacterial growth, anaerobic production was at least 30 times greater than aerobic culture and 50 and 100 times higher after the addition of albumin and haemoglobin. In addition, the apparent attenuation of yaminobutyric acid production by the addition of lactulose to albumin in the culture medium under aerobic conditions was not seen under anaerobic conditions or when production was related to the luxurience of bacterial growth. This cannot be explained by the lack of lactulose metabolism by E coli since its concentration was found to be reduced by 50 to 60% at the end of culture period. The concentrations of blood, albumin, and lactulose were chosen to cover the range of protein and lactulose concentrations likely to be found in the colon in patients with hepatic encephalopathy and are comparable with the concentrations used in previous studies4 which showed an effect on bacterial metabolism. However, in vivo studies by Conn et al23 showed no quantitative effect by lactulose on the colonic flora in normal subjects or patients with liver cirrhosis. There was excellent correlation between values obtained by the radioreceptor assay and by gas chromatography mass spectrophotometry. This confirms that the substance produced by E coli is true y-aminobutyric acid and also confirms the usefulness of the simpler radioreceptor assay. This study shows that the colon is a potential source of true y-aminobutyric acid and that protein significantly enhances its production by faecal bacteria. These data suggest that lactulose does not exert its beneficial effect by attenuating bacterial y-aminobutyric acid production. This does not necessarily militate against yaminobutyric acid having a role in the mediation of hepatic

encephalopathy as little is known about its utilisation within the colon by other bacteria or its absorption, both ofwhich could be affected by treatment with lactulose.

Effects of lactulose and lactitol on protein digestion and metabolism in conventional and germ free animal models: relevance of the results to their use in the treatment of portosystemic encephalopathy. / Bird SP, Hewitt D, Ratcliffe B, Gurr MI. // Gut. 1990 Dec;31(12):1403-6. AFRC Institute of Food Research, Laboratory, Shinfield. Protein digestion and metabolism have been studied in laboratory rats and miniature pigs to investigate the mechanisms of action of lactulose and lactitol when used in the treatment of patients with portosystemic encephalopathy. Lactulose (beta-D-galactopyranosyl-(1----4)-betaD-fructofuranose) and lactitol (beta-D-galactopyranosyl-(1----4)-D-glucitol) increased the excretion of nitrogenous material in the faeces and decreased nitrogen excretion in the urine in a similar degree to that reported for human patients. In studies with germ free rats given lactulose no such effect was observed, suggesting that, for lactulose at least, these effects are mediated by the gut flora. Measurement of the alpha-, epsilon-diaminopimelic acid content of the faeces confirmed that the enhancement of faecal nitrogen was due to an increased contribution from bacteria. The similarity in the results for lactulose and lactitol suggests that, from the perspective of protein metabolism, lactitol acts in a similar way to lactulose in the treatment of portosystemic encephalopathy. Lactulose (13-D-galactopyranosyl-(1-*4)-,3-Dfructofuranose) has been used in the treatment of portosystemic encephalopathy, a common complication of cirrhosis of the liver, ever since the pioneering experiments of Bircher et al.' Its action is not fully understood, however, but several hypotheses have been suggested, all of which focus mainly on events in the lower gut. Conn and Lieberthal have listed five theories2: bacteria ferment lactulose to organic acids, and the acidity produced encourages a more favourable gut flora; this acidity, and the osmotic effects of lactulose, have beneficial cathartic effects; the acidity inhibits production of ammonia by bacteria; the acidified luminal contents trap ammonia; lactulose stimulates bacterial growth and thereby the incorporation of luminal ammonia, and possibly ammonia from body tissue, into bacterial protein. There is evidence from animal studies that any carbohydrate which reaches the lower gut has a considerable stimulatory effect on bacterial activity and growth, and hence on the quantity and type of faecal nitrogen.3 Lactulose is not hydrolysed by human gut enzymes4 and as it is not absorbed in the small intestine2 it reaches the lower gut. In patients with cirrhosis there is increased faecal nitrogen output when lactulose is given5 and this may be important in understanding its action. Mason has suggested that the use of lactulose in the treatment of portosystemic encephalopathy may depend on it providing an energy source for bacteria in the lower gut, and he suggested some alternatives which may be superior.6 Such bacterial growth uses the nitrogenous components of the luminal contents which would otherwise be absorbed into the systemic circulation, overloading the body's capacity to metabolise them, and leading to the neurological effects that characterise portosystemic encephalopathy.2

Lactulose syrup has to be prescribed in such large amounts (30-150 ml or more, equivalent to 20100 g lactulose) that it can be regarded as a dietary supplement rather than a drug, but it is by no means ideal as a food ingredient. Substances with good functional properties would be attractive to doctors and the recently introduced crystalline lactulose may be preferable.7 There has been considerable interest in the use of lactitol (,B-D-galactopyranosyl-(1-+4)-D-glucitol) as an alternative for treating portosystemic encephalopathy because it is palatable and it may be as effective as lactulose syrup.8-'2 We have recently reported nutritional balance studies using the laboratory rat and the Gottingen miniature pig in which the energy values oflactulose and lactitol were determined. 13 A further objective of that work was to establish the precise effect of lactulose in the diet on the routes of excretion of nitrogen since Conn and Lieberthal2 suggested that this might explain its efficacy in the treatment of portosystemic encephalopathy. The studies on the laboratory rat were extended to elucidate the role of the gut flora. The effect of lactulose syrup on the bacterial component of faecal mass was assessed from the quantity of a-,ediaminopimelic acid in the faeces. This precursor of lysine is not present in animals and is only found in appreciable quantities in some bacteria. Thus, this unusual amino acid can serve as a crude indicator of bacterial mass. In a further experiment the effect Clearly, if the mechanism of action does involve the gut flora then lactulose should be without effect in the absence of bacteria. Methods STUDIES WITH LABORATORY RATS AND MINIATURE PIGS Conventional miniature pigs of the Gottingen strain, aged 11 weeks, and laboratory rats, aged 6 weeks, from the laboratory colony of barriermaintained, conventional Lister Norwegian hooded rats, were used in nutritional balance experiments. In one experiment 6 week old germ free and conventional rats of the Fisher strain were used. Details of procedures with conventional animals have been published elsewhere.'3 The same type of metabolism cage was used for conventional and germ free rats, the latter inside Trexler type flexible plastic film isolators.'4 All the rats were derived from the laboratory's small breeding colony of germ free animals, half of each litter being removed from the isolators at weaning to an animal room where they were conventionalised by dosing with a suspension of faeces collected from rats from the laboratory colony. In all experiments a cross over design with two periods was used in which each animal received the following dietary treatments. a control diet, without lactitol or lactulose syrup, and an experimental diet including one of these supplements. The animals were assigned to each of two groups by a random process. One group received the experimental diet in the first period and the control diet in the second; the other group was given the diets in the reverse order. DIETS The pigs were fed on a normal pig diet whereas the rats were given a semipurified diet based on casein and maize starch.'3 The levels of lactulose and lactitol tested varied as shown in Table I. Lactitol was incorporated in the control diet for rats at the expense of maize starch at the level of 100 g/kg. For pigs, the same amount of lactitol (100 g) was added to each kg of control diet. Rats would not tolerate this amount of lactulose and thus 75 gfkg was used. Pigs tolerated lactulose in

the diet and animals were given as much as they would take while still producing semisolid faeces that could be collected separately from the urine. NUTRITIONAL BALANCE STUDIES During the balance periods complete collections of urine and faeces were made corresponding accurately to particular feeding periods. Urine was collected into 005 M sulphuric acid to reduce losses of volatile nitrogenous substances. Samples were freeze dried before further processing.' 3 All faeces and urine collections were analysed for nitrogen while the collections from the first group of conventional rats given lactulose were also analysed for a-,r-diaminopimelic acid. ANALYTICAL METHODS Nitrogen was determined by a micro-Kjeldahl method. To determine a-,E-diaminopimelic acid, protein hydrolysis was followed by purification of the acid on an anionic exchange resin and colorimetric estimation after reaction with ninhydrin.'5 STATISTICAL ANALYSIS Data were subjected to standard analysis of variance for a cross over experiment'6 and the standard errors of mean values presented are based on the residual error mean squares with n-2 degrees of freedom where n is the number of animals. The experiment with germ free animals was a cross over design in each environment. In the statistical analysis the error for comparing diets within an environment was based on residual variation within animals with six degrees of freedom, whereas for comparing environments within diets a pool of this error, and the error between animals within environments with five degrees of freedom, was used. Results The effects of including lactulose syrup and lactitol in the diets of pigs and rats are shown in Table I. The amount of nitrogen excreted per unit nitrogen consumed is given. This form of standardisation was done to remove the effect of differences in the protein content of the diet on nitrogen excretion thus facilitating comparison between species and with published data for humans. Both supplements had consistently positive effects on the amount of nitrogen excreted in the faeces and negative effects on urinary nitrogen. The effects varied in degree, nitrogen in faeces urine being reduced by 7% to 34%. All effects were significant apart from lactitol on urinary nitrogen in the miniature pig. Table II shows the amounts of the bacterial amino acid a-,e-diaminopimelic acid in the faeces of rats on the control and the lactulose containing diets. With lactulose syrup in the diet the concentration in the faeces was about five times greater, and the total amount excreted was increased about six times compared with the control value. In an experiment on germ free and conventional rats the effects of adding lactulose syrup to the diet of the conventional animals (shown in Table III) were similar to those previously observed (Table I), though the absolute values were higher. In the germ free rats, however, the effect on faecal nitrogen was completely reversed and a significant interaction between diet and

environment was clearly established. In contrast, lactulose syrup had no effect on the nitrogen excreted in the urine by germ free animals.

Discussion Currently the standard treatment for portosystemic encephalopathy is 100 g lactulose/day, or roughly 160 g/kg of food dry matter. Our criterion for giving rats and pigs the doses used in these studies was the same as a doctor's - that is, maximum intake without precipitating diarrhoea. The amount given varied from 160 to 310 g/kg of dry matter or up to twice the amount a patient may consume. The maximum dose we used is not vastly greater than the human dose and appeared to be physiologically equivalent. Unfortunately, lactulose has been available only as a sickly sweet syrup that is rather disliked. Nevertheless, it has proved reliably successful. A complete understanding of how it works is desirable since such knowledge could offer the prospect of developing suitable alternatives of greater acceptability to patients. The first effect after consuming lactulose is probably catharsis, and indeed it is used successfully to treat constipation. Lactulose reaches the lower gut unchanged where it is metabolised by bacteria to short chain fatty acids'7 with a consequent lowering of pH'8 and thereby having a further cathartic effect. It would be expected that lactulose consumption would lead to considerable changes in the bacterial flora of the lower gut, but despite much study the picture is confused.2 The current work was done in normal healthy animals to clarify one aspect of the physiological effects of lactulose - that is, the effect on the excretion of the nitrogenous end products of protein digestion and metabolism. In our studies dietary supplements of both lactulose and lactitol increased the amount of nitrogen in the faeces and there was a concomitant, and generally

greater, reduction in urinary nitrogen. In Weber's patients lactulose increased faecal nitrogen by 125 mg/g food nitrogen and urinary nitrogen was decreased by 85 mg/g food nitrogen. 5 Thus in these respects the normal healthy rat and pig responded similarly to portosystemic encephalopathy patients. Evidence that giving lactulose involved the gut flora was obtained by estimating faecal bacteria by measuring a-,e-diaminopimelic acid. This amino acid has been used as a marker for bacterial nitrogen in studies of ruminant nutrition'9 since it is present only in bacteria, though not in all species, and not in protozoa and plant material. (Rumen bacteria contained 4-6 mg a-,ediaminopimelic acid/g dry cells.'9) Giving lactulose syrup to rats led to a considerable increase in the amount of this amino acid excreted, suggesting an increase in the number of bacteria in the faeces. The ratio of a-, c-diaminopimelic acid to nitrogen in bacteria varies widely, however, and it is possible that a treatment effect could arise from an alteration in the composition of the bacterial flora rather than from a change in the total number of bacteria. Nevertheless, the effect observed was so pronounced that it seems likely that there was an increase in the bacterial content of the faeces. Recently, Weber and coworkers assessed the effect of lactulose on faecal material directly.20 They applied a physical method of quantitative separation to patients' faeces and showed that lactulose increased both the bacterial nitrogen and, surprisingly, the non-bacterial soluble nitrogen components of the excreta. The effect on the bacteria confirms our results, though the experiments are not directly comparable due to the possible effects of the different diet types used.

The experiment with germ free rats provided further support for the involvement of the gut flora. It sould be pointed out, however, that there are differences between germ free and conventional animals which should be borne in mind in interpreting comparative results. Most importantly, there is the peculiar enlargement of the caecum and lower gut and the related physiological characteristics of germ free rodents. Differences in the efficiency ofprotein utilisation have been reported before, and we found that germ free rats on the control diet lost about a third more nitrogen than conventional rats. Similar observations, of a 22% increase21 and of a 20-50% increase (Z Ofuya et al, personal communication) have been made; these effects may be due to the absence of the deaminating activity of bacteria in the germ free animals. In contrast to the effect of lactulose in conventional animals this treatment caused an appreciable decrease in faecal nitrogen in germ free rats, suggesting the possibility of some counteraction of the effect of the lack of a gut flora. The absence of any positive effect on faecal nitrogen supports the idea that the effect of lactulose in conventional animals is mediated through the gut flora.

These results support the hypothesis that lactulose is effective in the treatment of portosystemic encephalopathy because it stimulates bacterial growth in the lower digestive tract. This conclusion may, however, be simplistic. The bacterial population in the gut is extremely complex in type and number so the result of any treatment is likely to be complicated. Our results do not exclude the possibility that other treatments that act by means of the gut flora, such as antibiotics, may bring additional benefit above that due to undigested carbohydrates such as lactulose. Antibiotics rarely remove all bacteria as is the case in the germ free animal, and in patients treated with appropriate antibiotics there will often remain scope for treatment with indigestible carbohydrates. The clinical evidence for additive effects of one such antibiotic, neomycin, and lactulose is equivocal.22 The gut flora can differ substantially between individuals and so too does individual tolerance to substances like lactulose. It seems reasonable to suppose, however, that appropriate antibiotics may usefully supplement lactulose treatment when substantial numbers of 'harmful' bacteria are unaffected by treatment with lactulose or when the patient will not consume enough lactulose for complete control of the metabolic activity of the gut flora. It seems that lactitol has a similar effect to lactulose syrup. Further work is required to confirm that these sugars act simply as energy sources for the growth of bacteria in the gut and do not have other specific actions that depend on their chemical structure.

Effects of lactulose on the intestinal microflora of periparturient sows and their piglets. / Krueger M, Schroedl W, Isik W, Lange W, Hagemann L. // Eur J Nutr. 2002 Nov;41 Suppl 1:I26-31. Institute for Bacteriology and Mycology, Veterinary Faculty, University of Leipzig, An den Tierkliniken 29, Germany. The periparturient period of animals (and humans) is very stressful and influenced by the microecosystem of the gastrointestinal tract (GIT). Performance and productivity of animal husbandry depend on the health of animal mothers and their offspring. We investigated the influence of prebiotic amounts of lactulose in sows and their piglets. Two experimental trial sows received daily 30 ml lactulose, 71 field trial sows received daily 45 ml lactulose during their periparturient period (10 days before until 10 days after parturison). The weaners of trial sows received 15 ml lactulose per 1 kg baby food 10 days before and 10 days after weaning.The effect of lactulose was recorded by performance parameters like number of piglet born alive, losses until weaning, body mass of piglets, daily weight gain of weaners until 35 days after weaning. The effect of lactulose on GIT microflora was estimated by bacterial counts of faeces of sows (total aerobic bacteria, Gram-negative bacteria, Clostridium (C.) perfringens). In order to show a previously unknown effect of lactulose we investigated the levels of antibodies to phospholipase C (PLC) of C. perfringens in plasma of experimental sows and in colostral and ripe milk of field sows. Lactulose influenced the performance parameters of sows in a nonsignificant way. In case of weaners we recorded significant daily weight gains. Lactulose significantly influenced total aerobic bacterial counts, C. perfringens counts in faeces of sows 20 days after parturison. Under experimental conditions it was shown that trial sows and their piglets had higher IgG-antibody levels to C. perfringens PLCs than the control animals. Similar

results were found under field conditions. Trial sows had significant higher IgG-anti LPS (J5) antibodies in milk 10 days after birth.

Lactulose feeding lowers cecal densities of clostridia in piglets. / Kien CL, Blauwiekel R, Williams CH, Bunn JY, Buddington RK. // JPEN J Parenter Enteral Nutr. 2007 MayJun;31(3):194-8. Department of Pediatrics, College of Medicine, University of Vermont, Burlington, Vermont, USA. cl.kien@uvm.edu BACKGROUND: In order to understand the consequences of persistent enteral feeding in patients with carbohydrate malabsorption, we fed piglets lactulose in sufficient dosage to produce osmotic diarrhea or inulin, using a conventional dose, to determine if this prebiotic can modulate the effects of lactulose. Feeding lactulose increases cecal luminal synthesis of butyrate, with inulin having an intermediate effect. Because clostridia may be a major source of colonic butyrate production, we hypothesized that feeding piglets lactulose or inulin would increase cecal densities of clostridia. METHODS: Piglets were assigned to 3 formula study groups for 6 days: (1) control, fed only sow milk replacer (n = 12); (2) inulin, inulin supplement (3 g/L; n = 11); and (3) lactulose, lactulose supplement (66.7 g/L; n = 6). Cecal fluid for bacteriological studies was sampled intraoperatively. RESULTS: The wet/dry ratio of the cecal contents (mean +/- SEM) was 8.2 +/- 0.5, 6.2 +/- 0.5, and 18.8 +/- 5.5, respectively, in the control, inulin, and lactulose groups (p = .049, KruskalWallis). There were no differences among the diet groups for cecal densities (10(6) colonyforming units [CFU]/g dry wt cecal contents) of total anaerobes, total aerobes, bifidobacteria, or lactobacilli. Densities of clostridia were markedly reduced in the lactulose group (1.14 +/- 0.41) vs the control (18.39 +/- 4.44; p = .001) or inulin groups (8.87 +/- 2.20; p = .04). CONCLUSIONS: In piglets, feeding lactulose at a dose known to cause diarrhea reduces cecal densities of clostridia.

4.2.1.2. . Lactulose: an indirect antioxidant ameliorating inflammatory bowel disease by increasing hydrogen production. / Chen X, Zuo Q, Hai Y, Sun XJ. // Med Hypotheses. 2011 Mar;76(3):325-7. Epub 2010 Oct 30. Graduates Management Unit, Second Military Medical University, Shanghai 200433, PR China. Lactulose, which cannot be digested and absorbed by body, is clinically widely used to treat constipation and hepatic encephalopathy. Fermented by gastrointestinal tract bacteria, lactulose can produce considerable amount of hydrogen, which is protective for DSS-induced colitis as a unique antioxidant. We propose that lactulose is an indirect antioxidant that mobilizes endogenous hydrogen production which in turn can reduce oxidative stress and ameliorate symptoms of inflammatory bowel disease in human beings.

Protective role of lactulose in intestinal carcinogenesis. / Hennigan TW, Sian M, Matthews J, Allen-Mersh TG. // Surg Oncol. 1995 Feb;4(1):31-4. Department of Surgery, Charing Cross and Westminster Medical School, London, UK. Primary bile acids are converted to carcinogenic secondary bile acids by colonic bacteria when the colonic pH is high. Therefore acidification of the luminal contents may reduce the cancer risk. The ability of lactulose to reduce colonic pH, secondary bile acid production, mucosa crypt cell production rate and tumour formation was measured in a rat model of intestinal carcinogenesis. Eighty Wistar rats were divided into four groups receiving normal diet alone or with lactulose, azoxymethane or both azoxymethane and lactulose. The addition of lactulose was associated with a significant fall in small intestinal and colonic pH. Lactulose was associated with a sharp rise in the secondary to primary bile acid ratio. The crypt cell production rate fell significantly with lactulose. The addition of lactulose was associated with a significantly reduced tumour yield in small intestine but not in colon. Lactulose therefore can reduce this is of no value in humans at risk of developing colorectal carcinoma.

Protective role of faecal pH in experimental colon carcinogenesis. / S. L. Samelson, R L Nelson, and L M Nyhus. // J R Soc Med. 1985 March; 78(3): 230233. There is epidemiological evidence that populations with alkaline stool pH are at greater risk for colon cancer than populations with acid stool pH. This association was investigated in the laboratory using the rat-dimethylhydrazine colon carcinogenesis model. Rats with acid stool pH, produced by consumption of lactulose or sodium sulphate or both, had significantly fewer colon tumours after injections of dimethylhydrazine (DMH) than rats treated with DMH alone. The results confirm the hypothesis that acidification of the stool can protect against the induction of colon cancer. Introduction

It has been hypothesized that alkaline faecal pH increases the risk of colon cancer (Thornton 1981, Burkitt 1981). We have used the rat-dimethylhydrazine colon carcinogenesis model to test this hypothesis. In rats, the injection of 1, 2-dimethylhydrazine (DMH) induces colon cancers and polyps identical to those found in humans. Stool pH can be lowered directly, without first altering bile flow, bile composition or colonic bacterial flora, through the feeding of lactulose, sodium sulphate or a combination of the two. We therefore set out to determine if fewer tumours were induced in rats given dietary supplements of lactulose or sodium sulphate or both and injections ofDMH than were induced in rats treated with DMH alone. Methods Twenty-five male Sprague-Dawley rats 4-6 weeks of age were assigned to each of four dietary regimens: rat chow alone, lactulose-supplemented rat chow, sodium sulphate-supplemented rat chow, and rat chow with both additives. On an intake-for-body-weight basis, the additives were provided in concentrations similar to those used in humans: 1.5 ml lactulose syrup or 50 mg sodium sulphate per 20 g pellet. After four weeks of acclimation to this diet, a series of 16 weekly subcutaneous injections ofDMH base, 15 mg/kg, were given to all rats. Eight weeks after the last injection, all rats were killed with an overdose of ether. The number, location and size of colon tumours were recorded, as was the presence of extracolonic tumours or metastases. Faecal pH was monitored with a Beckman model 41 pH meter and glass electrodes. Individual samples were emulsified with 1 cm3 of neutral 0.9% sodium chloride. In a pilot study, rats were killed and faecal pH was measured in the right and left colon. In the DMH study groups, freshly passed stool specimens were used. Statistical tests used included randomized analysis of variance, with the revised least significant difference method to discriminate between individual group means. The mean, variance and confidence limits were calculated for each group. Variables investigated included total number of tumours pe r rat, tumours per anatomic region of the colon, and a crude estimate of total tumour burden. Results The final groups varied in size from 21 to 23 rats. The total number of tumours per group was: regular diet alone without DMH, zero; DMH alone, 77; DMH plus lactulose, 55; DMH plus sodium sulphate, 53; and DMH plus both lactulose and sodium sulphate, 59. Colon cancer developed in all rats that received DMH. The data are summarized in Table 1. Results of pH measurements can be seen in Table 2. The total number of tumours was statistically different (P <0.05) between the DMH-alone group and all diet-supplemented groups. There was no statistical difference in tumour location, burden or metastases. Also the lactulose and sodium sulphate supplemented groups were statistically indistinguishable for all the variables measured.

Discussion The epidemiological association of high colon cancer risk with low consumption of crude fibre and high consumption of animal fat, animal protein and beer is well known. Other- major risk factors include a substantial genetic component, even when patients with known familial syndromes such as familial polyposis are excluded, and alkaline faecal pH. Details of the three reports summarizing the latter association are shown in Table 3 (MacDonald et al. 1978, Malhotra 1982, Pietroiosti et al. 1983). In our experiment we wished to determine if primary pharmacological alteration of stool pH would alter the number of colon tumours induced in rats. The results showed that acidification of the stool using either lactulose or sodium sulphate significantly reduced the number of colon tumours induced. It is worth stressing that acidification only of the faeces of these rats was demonstrated (throughout the colon as well as in the expelled faeces). It is not known if the surface of the colonic mucosa was also acidified or if the mucous barrier prevented this acidification. The mechanism by which an acid pH protects against colon cancer is unknown, though it may be decreased 7-alpha-dehydroxlation of bile salts by bacterial enzymes in an acid environment (Thornton 1981).

Two means were chosen by which to lower stool pH - lactulose and sodium sulphate. Lactulose is a synthetic, nonabsorbable disaccharide which is metabolized into organic acids by colonic bacteria. These acids reduce the pH of the right colon to as low as 4.5. Lactulose has been shown to have relatively little effect on the pH of the contents of the left colon. Sodium sulphate is a stimulant laxative that decreases stool pH by a mechanism different from that of lactulose. Its greatest effect is in the left colon. In humans, the best results of acidification of the contents of the entire colon were achieved when the two drugs were given together (Bown et al. 1974). In our experiment, similar degrees of acidification and protection were found regardless of the substance used to lower stool pH (Tables 1 and 2.) One objection that might be raised to the conclusions drawn from this experiment is that, since lactulose and sodium sulphate are both laxatives, the decreased induction of tumours was caused by decreased faecal transit time rather than reduced stool pH. There are several reasons why this may not be so. First, though dietary fibre was believed to exert its protective effect in colon cancer by decreasing faecal transit time, epidemiological studies that have measured transit time have not supported this theory. Japanese residents of Hawaii, who have a risk of colon cancer similar to that of Americans, maintain the shorter faecal transit time of the Japanese (Glober et al. 1977). In the rat-DMH model, supplementation in the diet of cellulose fibre did not confer any protection against colon tumour induction (Ward et al. 1973), nor did supplementation of psyllium seeds, a bulk laxative (Castleden 1977), nor magnesium sulphate, a stimulant laxative that does not lower stool pH (Cleveland & Cole 1969). Last, interposition of a short segment of colon in the rat into the proximal jejunum, where transit time would be extremely rapid, again provided no protection in the transposed segment against tumours induced by DMH (Celik et al. 1981). Therefore, the decreased induction of colon tumours in the rats given dietary supplements that reduced stool pH is most likely to have resulted from colon acidification and not from decreased faecal transit time. A second point to address is whether or not alteration of stool pH might have resulted in altered bile flow or bacterial flora, since changes in either of these factors are known to have an effect on colon cancer risk (Nelson 1983). Acidification of the stool in humans has been shown to alter the chemical composition of the bile and the degree of saturation of the bile with cholesterol (Thornton 1981, Thornton & Heaton 1981). Some factors that relate hepatic metabolism to colon cancer risk include epidemiological evidence that patients who have had a cholecystectomy have an increased risk for colon cancer (Linos et al. 1981). Cholecystectomy has been shown to increase the incidence of tumours of the colon in mice treated with DMH (Werner et al. 1977). It is certainly important when studying the complex relationship of hepatic and colonic metabolism to remember that the liver and the colon are connected by a two-way interaction. Once again, we do not know the complete mechanism by which our dietary

supplement decreased the number of tumours induced, only that the first step of the mechanism was acidification of the contents of the rats' colons. The fact that both supplements had equivalent effectiveness in acidification and tumour reduction by different mechanisms strengthens this relationship. How might stool pH be lowered in humans? Wheat bran is quite similar to lactulose in that it is metabolized by colonic bacteria to short-chain fatty acids and thereby acidifies the stool (Cummings et al. 1976, Thornton 1981). Bran also has an effect similar to that of lactulose on the chemical composition of bile (Pomare et al. 1976). It is probably by these mechanisms that bran consumption decreases colon cancer risk rather than the alteration of faecal transit time (Burkitt 1981). Another means of lowering stool pH for individuals who are lactase-deficient (most Orientals, 70% of Blacks, and 30% of Caucasians) is the consumption of dairy products. Undigested lactose in the colon will act just as lactulose does and so, at the risk of some flatulence, diarrhoea and occasional upset stomach, may provide protection against colon cancer.

4.2.1.3. .

4.2.1.4. .

4.2.2. .

4.2.2.1. . .

4.2.2.2. . Poorly absorbed from GI tract. Less than 3% absorbed from small intestine following oral administration; negligible absorption from colon.

4.2.2.3. . .

4.2.2.4. . Absorbed drug not metabolized. Unabsorbed drug reaches colon unchanged, where it is metabolized by bacteria to form lactic acid and small amounts of acetic and formic acids.

4.2.2.5. . Absorbed drug excreted in urine unchanged within 24 hours.

4.2.2.6. (). .

4.2.2.7. . .

4.2.3. .

4.2.3.1. . The acute oral LD50 of the drug is 48.8 mL/kg in mice and greater than 30 mL/kg in rats.

4.2.3.2. . .

4.2.3.3. . There are no known animal data on long-term potential for mutagenicity.

4.2.3.4. Administration of Lactulose solution in the diet of mice for 18 months in concentrations of 3 and 10 percent (v/w) did not produce any evidence of carcinogenicity.

4.2.3.5. . In studies in mice, rats, and rabbits, doses of Lactulose solution up to 6 or 12 mL/kg/day produced no deleterious effects on breeding, conception, or parturition. Reproduction studies have been performed in mice, rats, and rabbits at doses up to 2 or 4 times the usual human oral dose and have revealed no evidence of impaired fertility or harm to the fetus due to Lactulose.

4.2.3.6. . .

4.2.3.7. . .

4.3. .

Effect of lactulose on establishment of a rat non-alcoholic steatohepatitis model. / Fan JG, Xu ZJ, Wang GL. // World J Gastroenterol. 2005 Aug 28;11(32):5053-6. 2. Comparison of probiotics and lactulose in the treatment of minimal hepatic encephalopathy in rats. / Jia L, Zhang MH. // World J Gastroenterol. 2005 Feb 14;11(6):908-11. 3. Effect of lactulose on short-chain fatty acids and lactate production and on the growth of faecal flora, with special reference to Clostridium difficile. / Ito Y, Moriwaki H, Muto Y, Kato N, Watanabe K, Ueno K. // J Med Microbiol. 1997 Jan;46(1):80-4. 4. Effect of protein and lactulose on the production of gamma-aminobutyric acid by faecal Escherichia coli. / al Mardini H, al Jumaili B, Record CO, Burke D. // Gut. 1991 Sep;32(9):1007-10. 5. Effects of lactulose and lactitol on protein digestion and metabolism in conventional and germ free animal models: relevance of the results to their use in the treatment of portosystemic encephalopathy. / Bird SP, Hewitt D, Ratcliffe B, Gurr MI. // Gut. 1990 Dec;31(12):1403-6. 6. Effects of lactulose on the intestinal microflora of periparturient sows and their piglets. / Krueger M, Schroedl W, Isik W, Lange W, Hagemann L. // Eur J Nutr. 2002 Nov;41 Suppl 1:I26-31. 7. Lactulose feeding lowers cecal densities of clostridia in piglets. / Kien CL, Blauwiekel R, Williams CH, Bunn JY, Buddington RK. // JPEN J Parenter Enteral Nutr. 2007 MayJun;31(3):194-8. 8. Lactulose: an indirect antioxidant ameliorating inflammatory bowel disease by increasing hydrogen production. / Chen X, Zuo Q, Hai Y, Sun XJ. // Med Hypotheses. 2011 Mar;76(3):325-7. Epub 2010 Oct 30. 9. Protective role of lactulose in intestinal carcinogenesis. / Hennigan TW, Sian M, Matthews J, Allen-Mersh TG. // Surg Oncol. 1995 Feb;4(1):31-4. 10. Protective role of faecal pH in experimental colon carcinogenesis. / S. L. Samelson, R L Nelson, and L M Nyhus. // J R Soc Med. 1985 March; 78(3): 230233.

1.