European Journal of Clinical Nutrition (2006) 60, 847–852 & 2006 Nature Publishing Group All rights reserved 0954-3007/06 $30.00 www.nature.com/ejcn ORIGINAL ARTICLE 15 N-excretion of heat-killed Lactobacillus casei in humans KD Wutzke and V Sattinger Children’s Hospital, Research Laboratory, University of Rostock, Rostock, Germany Objective: In the present study, Lactobacillus casei (L. casei) labelled with 15N was used to follow the metabolic fate of orally administered heat-killed 15N-labelled L. casei (15N-L.casei) in humans. Design: Experimental study. Setting: University of Rostock, Children’s Hospital, Research Laboratory. Subjects: Twelve healthy adults aged 23–32 years. Intervention: The subjects received 36 mg/kg body weight heat-killed 15N-L.casei and 500 mg Lactose-[13C]ureide together with breakfast. Expired air samples were taken over 14 h, whereas urine and faeces were collected over 2 days. A blood sample was taken after 2 h. 13C- and 15N-enrichments were measured by isotope ratio mass spectrometry (SerCon, UK). Results: The orocaecal transit time (OCTT) was reached after 4.1 h. The urinary 15N-excretion was 9.3% of the ingested dose, whereas the faecal excretion was 65.1% of the ingested dose. After 2 h, 15N-enrichment of supernatant, fibrinogen, and plasma protein precipitate amounted to 254, 11, and 2 p.p.m. excess, respectively. Conclusions: In comparison to the OCTT of 4.1 h, 15N-enrichment in urinary ammonia and urinary total nitrogen already began to rise 30 min after 15N-L.casei ingestion, indicating that 15N-L.casei is rapidly digested in the small bowel. This is confirmed by 15N-enrichments of blood plasma fractions. The ingestion of heat-killed 15N-L.casei led to a total excretion of 74.4% of the ingested 15N-dose. European Journal of Clinical Nutrition (2006) 60, 847–852. doi:10.1038/sj.ejcn.1602389; published online 1 February 2006 Keywords: 15 N-labelled Lactobacillus casei; 15 N-excretion; lactose-[13C]ureide; orocaecal transit time Introduction Lactobacillus casei (L. casei) is a Gram-positive, facultative anaerobic member of the commercially important lactic acid bacteria and possesses a strictly fermentative metabolism with lactic acid as the major metabolic end product (Kandler and Weiss, 1986; Axelsson, 1998). L. casei is commonly found in probiotic dairy foods such as viable yoghurts. L. casei also complements the growth of Lactobacillus johnsonii (La1). For development of potential health benefits to the consumer, the consumed yoghurt bacteria need to be Correspondence: Professor KD Wutzke, Children’s Hospital, Research Laboratory, University of Rostock, Rembrandtstrasse 16/17, D-18055 Rostock, Germany. E-mail: [email protected] Guarantor: KD Wutzke. Contributors: VS was principal investigator and doctoral candidate of KDW. Received 26 June 2005; revised 11 November 2005; accepted 2 December 2005; published online 1 February 2006 viable. Nevertheless, comparisons have been made between viable and dead probiotic bacteria reported in various studies assessing the clinical efficacy of heat-killed lactobacilli, for example, in the treatment of diarrhoea (Xiao et al., 2003), in the decrease of Helicobacter pylori colonisation (Cruchet et al., 2003) or in the oral rehydration therapy (Simakachorn et al., 2000). However, the metabolic fate of dead lactobacilli remained unknown. In a recently published study, we investigated the metabolic fate of viable La1 doubly labelled with 13C and 15N after oral ingestion in humans. Our findings showed that 50% of both stable isotopes were excreted in breath, urine, and faeces. In continuation of these investigations it was of interest to evaluate the effect of heat killing of Lactobacilli when used for oral ingestion. The aim of the present study was to follow the metabolic fate of orally administered heat-killed 15N-labelled L. casei (15N-L.casei) during the passage through the human intestine in correlation to orocaecal transit time (OCTT). 15 N-excretion of Lactobacillus casei KD Wutzke and V Sattinger 848 Methods Preparation of 15N-yeast extract The preparation of 15N-labelled Saccharomyces cerevisiae yeast from 15NH4Cl (99.0 atm%, Campro Scientific, Berlin) has been previously described in detail (Wutzke et al., 1983, 1992). A total of 125 g (wet weight) of 15N-labelled yeast cells suspended in 1 l distilled water were killed and hydrolysed by heating at 1001C for 45 min. The remaining cells were removed by centrifugation. The supernatant autoclaved at 1211C for 15 min was used as 15N-yeast extract (Table 1). Culture of lactobacilli L. casei lactobacilli were obtained from ‘Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH’, Braunschweig, Germany. The organisms were inoculated into 50 ml PY-bouillon (peptone yeast extract medium) and incubated at 371C for 2 days. All ingredients except vitamins and ascorbic acid were autoclaved at 1211C for 15 min. Solutions of vitamins and ascorbic acid were sterilised separately by filtration. After mixture of the filtrated and autoclaved solutions, the medium was adjusted to pH 6.8 with 10% NaOH (Wutzke and Oetjens, 2005). After incubation, 10 ml of L. casei containing PY-bouillon were centrifuged at 4000 r.p.m. The resulting sediment suspended in 5 ml 0.9% NaCl solution was used as an inoculum (8 109 colony forming units (cfu)) for 1 l of 15 N-L.casei broth medium (Table 1). After incubation at 371C for 72 h, 15N-L.casei was separated from the culture medium by centrifugation at 4000 r.p.m. and was purified by washing 4 times with 0.9% NaCl solution. The viable microorganisms were centrifuged at 14 000 r.p.m. Thereafter, the cell sediment was heated for 15 min at 1001C in an autoclave. The N-content in the aqueous phase was measured by using an elemental analyser. The N-extraction caused by the heatkilling procedure was found to be merely 0.7%. The microscopic examination of the gram stain preparation showed approximately 95% intact cell walls. Therefore, they Table 1 Composition of 15 N-yeast extract Glucose Ascorbic acid K2HPO4 3 H2O MgSO4 7 H2O MnSO4 2 H2O NaCl FeSO4 7 H2O Ca-pantothenate Biotin Tween-80 B vitamin mixture Distilled water ad L. casei European Journal of Clinical Nutrition 15 N-labelled Lactobacillus casei broth medium 1.0 l 50.0 g 50.0 g 12.5 g 0.5 g 50.0 mg 50.0 mg 25.0 mg 1.0 mg 0.25 mg 5.0 ml 0.5 ml 5.0 l o1.0 mg were finally stored without centrifugation at 201C in a freezer. Reactivating tests of the heat-killed 15N-L.casei cells failed on both agar plates and PY-bouillon. Subjects Twelve healthy adults (five female, seven male, mean age: 26.2 years, mean body weight: 67.5 kg, mean body mass index: 26.2 kg/m2), volunteered for this study. None of the subjects were receiving any medication or had a history of gastrointestinal diseases. All subjects were in good health throughout the study and none of the subjects complained about any gastrointestinal problems. The testing protocol was approved by the Committee on Ethics of the Faculty of Medicine of the University of Rostock. Study protocols Subjects were studied after an overnight fast. All volunteers received a continental breakfast made up of two wheat flour rolls, honey, and coffee beginning at 0745 hours. All volunteers received an individually standardised regular lunch made up of two wheat flour rolls, tomato soup, and two small sausages at noon. No other food was allowed until 4 h after breakfast. Baseline breath, blood, urine, and faeces samples were collected at 0730 hours, 30 min before the ingestion of both 15N-L.casei and lactose-[13C]ureide (13C-LU) to determine the baseline abundance of 15N and 13C. The subjects received a predosage of 3 200 mg unlabelled lactose ureide at 1200, 1600 and 2000 hours on the day before the ingestion of 13C-LU (Wutzke et al., 1997; Gohr, 2004; Wutzke and Glasenapp, 2004). During breakfast, 15 N-L.casei and 13C-LU were administered simultaneously as a single oral pulse labelling at 0800 hours in a dosage of 35.8 mg 15N-L.casei per kg body weight and 500 mg 13C-LU per subject stirred in 125 g L. casei yoghurt (Danone). The unlabelled commercial yoghurt was heated for 15 min at 1001C before ingestion. Two expired breath samples obtained by direct exhalation into Exetainerst were collected in 30 and 60 min-intervals over a period of 14 h for measuring 13CO2-enrichment (Wutzke et al., 1997). Total urine volume was collected at 0.5, 1, 2, 4, 12 and 24h intervals over a period of 48 h after 15N-L.casei administration. The urine fractions were pooled between the time intervals. Faeces scraped from polyvinyl film were also collected quantitatively over 2 days. Sample preparation before isotope ratio mass spectrometry analysis The sample preparation before isotope ratio mass spectrometry (IRMS) analysis has been recently explained in detail (Wutzke and Oetjens, 2005). Urine volumes and faecal masses were homogenised, recorded and stored at 201C until analysis (Wutzke et al., 1992). Ammonia was separated 15 N-excretion of Lactobacillus casei KD Wutzke and V Sattinger 849 in vessels by micro diffusion, following alkalisation of the urine with NaOH and trapped in boric acid. (Faust et al., 1981; Wutzke et al., 1992). A blood sample was taken 2 h after 15N-L.casei administration at 1000 hours. After centrifugation of the citrated blood, the plasma was used for fibrinogen extraction with CaCl2 solution (Faust et al., 1981). For plasma protein precipitation, the remaining fibrinogen-free plasma was treated with Na2WO4. The soluble supernatant was removed by centrifugation at 4000 r.p.m. (Faust et al., 1981). Fibrinogen, plasma protein precipitation, and supernatant were stored at 201C until analysis (Wutzke and Oetjens, 2005). Analytical techniques Isotope ratio mass spectrometry. The solid (faeces) and liquid samples were burned in an elemental analyser (SL) at 9001C to CO2, H2O and total-15N2 before the 15N-measurement. The 15N-enrichments in 15N-L.casei, fibrinogen, plasma protein precipitate, supernatant, urine, faeces, as well as the 13C-enrichment in breath were measured by isotope ratio mass spectrometry (Tracer mass 20–20, SerCon, Crewe, UK) as detailed previously (Barrie et al., 1989; Wigger and Wutzke, 2004; Wutzke and Glasenapp, 2004; Wutzke and Oetjens, 2005). The data were expressed either as enrichment d 13C and d 15N over baseline (DOB), as atom percent excess (APE), as parts per million excess (p.p.m. exc) (Slater et al., 2001), or as percentage of cumulative urinary and faecal 15 N-excretion (PCE). Percentage cumulative urinary and faecal 15N-excretion The percentage cumulative 15N-excretion (PCE) is the product of the urinary or faecal 15N-excretion (E) in mmol, respectively, times 100 divided by the 15N-excess dose (D) in mmol, respectively. Results All results are quoted as mean7s.d. Yield of 15N-L.casei The yield of bacterial wet mass was 1 g/l broth medium (3 1011 cells). The 15N-enrichment of 15N-L.casei amounted to 82.2 atom %, respectively. Breath CO2-signals were observed in breath beginning 30 min after 13C-LU ingestion. The maximum 13CO2-enrichment of 10.4 DOB was reached after 9 h and was still detectable after 14 h (Figure 1). According to our definition, the measured OCTT was 4.171.7 h (Figure 1). 13 Urine and faeces Figure 2 shows the 15N-enrichment in urinary total-nitrogen (N) and ammonia. The mean maximum 15N-enrichments in total-N (0.019 APE) and ammonia (0.033 APE) were reached after 3 h. The urinary PCE shows that 9.372.1% of ingested dose was excreted 48 h after 15N-L.casei administration. Figure 3 shows that 1.0% (according to 10.9% of the total-15N recovery after 48 h) is renally excreted before reaching the OCTT, whereas 8.3% (according to 89.1%) are excreted after 4.1 h. The faecal 15N-excretion amounted to 65.1% of ingested dose, 2 days after 15N-L.casei ingestion. Thus, 74.4% of the ingested 15N-dose was excreted in urine and faeces. The numbers of L. casei cfu/g in wet faeces were not counted. Blood plasma fractions The measured 15N-excess-enrichments amounted to 254791 in supernatant, 11732 in fibrinogen, and 275 p.p.m. exc PCE ¼ E100=D: D ¼ 0.745 mmol 15N/g 15N-L.casei (wet weight) ¼ 0.0267 mmol 15N/kg body weight. OCTT For measuring the OCTT, we used the lactose-[13C]ureide 13 CO2-breath test by predosing with unlabelled lactose ureide (Heine et al., 1995; Wutzke et al., 1997; Wutzke and Glasenapp, 2004). In 2004 our Rostock group applied a new, simplified and cheaper predosing protocol. It could be demonstrated that the predosing with 3 200 mg unlabelled lactose ureide is high enough to induce the enzyme activity and to give reliable results (Gohr, 2004). The OCTT was calculated from the time interval between 13 C-LU intake and the detection of a significant and sustained 13CO2 rise in breath of two DOB or more (Wutzke et al., 1997). Figure 1 Mean 13CO2-enrichment in breath and the resulting orocaecal transit time after lactose-[13C]ureide ingestion. European Journal of Clinical Nutrition 15 N-excretion of Lactobacillus casei KD Wutzke and V Sattinger 850 Figure 2 Mean urinary ammonia. 15 N-enrichment in urinary total nitrogen, and Figure 3 Mean percentage cumulative renal correlation to the orocaecal transit time. 15 N-excretion in in plasma protein precipitates, respectively. The fibrinogen and plasma protein enrichments did not differ significantly. Discussion The human gastrointestinal tract (GIT) possesses a very complex population of microorganisms, known as gut microbiota, that interact with each other and with the host. Estimates assess the count of different types of microorganisms in the gut at 400 and the total number of bacterial cells at 1014; a dimension that far exceeds the total number of human beings on the earth. Probiotics are defined as viable microorganisms, including Lactobacillus species, such as L. johnsonii and L. casei that may beneficially affect the host upon ingestion by improving the balance of the intestinal microflora (Fuller, 1989; Lee and European Journal of Clinical Nutrition Salminen, 1995; Pfeiffer and Rosat, 1999; Oozeer et al., 2002). L. casei is an adaptive species, and can be isolated from raw and fermented dairy products as well as from the intestinal tract of humans (Kandler and Weiss, 1986). L. casei has an industrial application as a human probiotic and as an acid-producing starter culture for milk fermentation (Fox et al., 1998). During the passage through the GIT, viable and dead yoghurt bacteria are exposed to the acidic conditions of the stomach, followed by the hydrolytic enzymes and bile salts of the small intestine (Meydani and Ha, 2000). Mainville et al. (2005) investigated the in vitro digestion of Bifidobacterium infantis, La1, Lactobacillus rhamnosus and other strains by using a reactor model simulating stomach (pH 2, presence of pepsin) and duodenum (pancreatin or bile salts) conditions. They observed that a few strains were not able to survive in both reactors. Recently published findings revealed that approximately equal amounts of 13C and 15N were excreted in faeces (39.9 and 37.6%, respectively) after ingestion of doubly stable isotope labelled viable La1 lactobacilli in healthy adults, whereas the urinary excretion amounted to 1.3 and 12.4%, respectively (Wutzke and Oetjens, 2005). Thus, total excretion of both isotopes was approximately 50% of the ingested dose. However, the metabolic fate of orally ingested dead lactobacilli remains open. In vivo tests in humans aimed at the investigation of the digestion of heatkilled lactobacilli are of great interest with respect to some important findings of these probiotic bacteria (Simakachorn et al., 2000; Cruchet et al., 2003; Xiao et al., 2003). Therefore, in the present study we investigated the metabolic fate of heat-killed 15N-L.casei after oral administration in correlation to the OCTT. Owing to the relatively high costs for [U-13C6]glucose we refrained from the doubly labelling of L. casei. The 15N-L.casei cells and 13C-LU were given simultaneously during breakfast to represent the transit time of the meal. Furthermore, the 15N-L.casei was ingested together with heat-killed unlabelled L. casei yoghurt to obtain a better distribution in the meal. When considering the results obtained after 15N-L.casei and 13C-LU ingestion, the time lag between the 15N-increase in urinary total nitrogen (N) and urinary ammonia and the rise of 13CO2 in breath, respectively, became obvious: in comparison to the prompt 15 N-enrichment in urinary-N and ammonia 0.5 h after 15 N-L.casei administration, the 13CO2-onset appears approximately 3.6 h later, clearly indicating that 15N-L.casei is rapidly digested and absorbed in the small bowel and in the colon (Figures 1, 2 and 3). The corresponding prompt 15 N-enrichments of the blood plasma fractions evidenced that 15N was incorporated in the metabolic pool 2 h after 15N-L.casei ingestion. In comparison to the plasma protein precipitates, higher 15N-enrichments were measured in fibrinogen, a protein with a high turnover. However, the highest 15N-enrichment was observed in the 15N-urea, free 15 N-amino acids and other soluble 15N-substrates containing supernatant, indicating the release of 15N-amino acids after 15 N-excretion of Lactobacillus casei KD Wutzke and V Sattinger 851 15 N-L.casei digestion and absorption of its degradation products. Therefore, it can be assumed that a substantial portion of ingested 15N-L.casei is rapidly digested in the small intestine and that the metabolites are incorporated into human tissue. When comparing the data obtained after ingestion of doubly labelled viable La1, the differences to the present study with heat-killed 15N-L.casei become obvious: the faecal 15 N-excretion of heat-killed 15N-L.casei was clearly higher in comparison to the data derived from viable La1 (65.1 vs 37.6% of ingested dose, respectively) whereas the urinary 15 N-excretion was in the same order of magnitude 9.3 and 12.4% of ingested dose, respectively. Figure 3 shows that the major part of 15N is renally excreted after reaching the caecum. Nevertheless, due to the delayed renal 15N-excretion it is not possible to differentiate between small intestinal and colonic recovery from the data. Two days after 15N-L.casei ingestion, the 15N-enrichment in urinary-N had not reached the baseline level, indicating that additional 15N is renally excreted over time. It can be assumed that a part of the 15N is incorporated and recycled in the large protein pool of the human body. Therefore, the delayed 15N-excretion leads to an underestimation of the excreted 15N label and to an overestimation of the incorporated 15N label, respectively. In comparison to the data of the present study, the lower faecal 15N-excretion observed in our previous study with viable 13C-,15N-labelled La1 reflected the colonisation effect of the probiotic cells in the gut. The higher the part of viable cells is able to adhere to the intestinal mucosa the lower their faecal excretion. The delayed degradation of the viable cells adhering to the mucosa could lead to a prolonged 15 N-recycling, resulting in a similar or higher urinary 15 N-recovery from viable lactobacilli in comparison to heatkilled cells. Nevertheless, it is difficult to compare the 15 N-excretion data of both studies since two different strains have been used. Owing to the total combustion of 15 N-L.casei to 15N2 in the SL, the origin of the stable isotope 15 N excreted in faeces remains unknown. It may be excreted as intact dead cells or in form of metabolic degradation products of 15N-L.casei such as indigestible 15N-protein or 15 N-nucleic acids. The OCTT of 4.1 h measured by using 13C-LU was in agreement with the data of 3.7 h calculated by the raffinose H2 breath test (Wutzke and Oetjens, 2005). After oral ingestion of 15N-labelled bifidobacteria in infants similar results were observed. The renal and faecal 15N-excretion ranged between 9.2–17.4 and 10.8–19.6% of ingested dose, respectively (Heine et al., 1992). Our combination of measuring the urinary and faecal excretion of 15N-labelled metabolic degradation products of heat-killed 15N-L.casei in correlation to the OCTT is a novelty. Further studies with heat-killed doubly labelled La1 and viable 15N-L.casei are in preparation. Our findings will contribute to the understanding of the digestion and elimination of viable and dead lactobacilli when consumed in humans. References Axelsson L (1998). Lactic acid bacteria: classification and physiology. In: Salminen S, Von Wright A (eds). Lactic Acid Bacteria: Microbiology and Functional Aspects 2nd edn. Marcel Dekker Inc.: New York. pp 1–72. Barrie A, Davies JE, Park AJ, Workmann CT (1989). Continuous-flow stable isotope analysis for biologists. Spectroscopy 4, 42–52. Cruchet S, Obregon MC, Salazar G, Diaz E, Gotteland M (2003). Effect of the ingestion of a ditary product containing Lactobacillus johnsonii La1 on Helicobacter pylori colonization in children. Nutrition 19, 716–721. Faust H, Bornhak H, Hirschberg K, Jung K, Junghans P, Krumbiegel P et al. (1981). Klinisch-chemische und isotopenanalytische Methoden zur Untersuchung des Stickstoff-Stoffwechsels mit N-15 beim Menschen. In: Akademie der Wissenschaften (eds). Methodenkatalog: Leipzig. pp 1–210. Fox PF, McSweeney PLH, Lynch CM (1998). Significance of nonstarter lactic acid bacteria in cheddar cheese. Aust J Dairy Technol 53, 83–89. Fuller R (1989). Probiotics in man and animals. J Appl Bacteriol 66, 365–378. Gohr D (2004). The effect of the lactose ureide dosage on the measurement of the orocaecal transit time. Doctoral theses University of Rostock. pp 1–70. Heine WE, Berthold HK, Klein PD (1995). A novel stable isotope breath test: 13C-labelled glycosyl ureides used as noninvasive markers of intestinal transit time. Am J Gastroenterol 90, 93–98. Heine WE, Mohr C, Wutzke KD (1992). Host-microflora correlations in infant nutrition. Prog Food Nutr Sci 16, 181–197. Kandler O, Weiss N (1986). Genus lactobacillus. In: Sneath PHA, Mair NS, Sharpe ME, Holt JG (eds). Bergey’s Manual of Systematic Bacteriology, vol. II. Williams & Wilkins: Baltimore. pp 1063–1065. Lee YK, Salminen S (1995). The coming age of probiotics. Trends Food Sci Technol 6, 241–245. Mainville I, Arcand Y, Farnworth ER (2005). A dynamic model that simulates the human upper gastrointestinal tract for the study of probiotics. Int J Food Microbiol 99, 287–296. Meydani SN, Ha WK (2000). Immunologic effects of yogurt. Am J Clin Nutr 71, 861–872. Oozeer R, Goupil-Feuillerat N, Alpert CA, van de Guchte M, Anba J, Mengaud J et al. (2002). Lactobacillus casei is able to survive and initiate protein synthesis during its transit in the digestive tract of human flora-associated mice. Appl Environ Microbiol 68, 3570–3574. Pfeiffer A, Rosat JP (1999). Probiotics in alimentation: clinical evidence for their enhancement of the natural immunity of the gut. In: Hanson LA, Yolken RH (eds). Probiotics, Other Nutritional Factors and Intestinal Microflora, Nestlé nutrition workshop series vol. 42. Lippincott, Raven: Philadelphia. pp 243–255. Simakachorn N, Pichaipat V, Rithipornpaisarn P, Kongkaew C, Tongpradit P, Varavithya W (2000). Clinical evaluation of the addition of lyophilized, heat-killed Lactobacillus acidophilus LB to oral rehydration therapy in the treatment of acute diarrhea in children. J Pediatr Gastroenterol Nutr 30, 68–72. Slater C, Preston T, Weaver LT (2001). Stable isotopes and the international system of units. Rapid Commun Mass Spectrom 15, 1270–1273. Wigger M, Wutzke KD (2004). Whole-body protein turnover in nephrotic subjects: a comparison between different degrees of proteinuria. Isotopes Environ Health Stud 40, 155–161. Wutzke KD, Glasenapp B (2004). The use of 13C-labelled glycosyl ureides for evaluation of orocaecal transit time. Eur J Clin Nutr 58, 568–572. Wutzke KD, Heine W, Drescher U, Richter I, Plath C (1983). 15 N-labelled yeast protein – a valid tracer for calculating wholebody protein parameters in infants: a comparison between [15N]-yeast protein and [15N]-glycine. Hum Nutr Clin Nutr 37C, 317–327. European Journal of Clinical Nutrition 15 N-excretion of Lactobacillus casei KD Wutzke and V Sattinger 852 Wutzke KD, Heine WE, Plath C, Leitzmann P, Radke M, Mohr C et al. (1997). Evaluation of oro-coecal transit time: a comparison of the lactose-[13C, 15N]ureide 13CO2- and the lactulose H2- breath test. Eur J Clin Nutr 51, 11–19. Wutzke KD, Heine W, Plath C, Müller M, Uhlemann M (1992). Whole-body protein parameters in premature infants: a comparison of different 15N tracer substances and different methods. Pediatr Res 31, 95–101. European Journal of Clinical Nutrition Wutzke KD, Oetjens I (2005). 13C- and 15N-incorporation of doubly stable isotope labelled Lactobacillus johnsonii in humans. Eur J Clin Nutr 59, 1167–1172. Xiao SD, Zhang de Z, Lu H, Jiang SH, Liu HY, Wang GS et al. (2003). Multicenter, randomized, controlled trial of heat-killed Lactobacillus acidophilus LB in patients with chronic diarrhea. Adv Ther 20, 253–260.
© Copyright 2026 Paperzz