Yoghurt enriched with Lactobacillus acidophilus does not lower blood ...

European Journal of Clinical Nutrition (1999) 53, 277±280 ß 1999 Stockton Press. All rights reserved 0954±3007/99 $12.00 http://www.stockton-press.co.uk/ejcn

Yoghurt enriched with Lactobacillus acidophilus does not lower blood lipids in healthy men and women with normal to borderline high serum cholesterol levels NM de Roos1 *, G Schouten1 and MB Katan1;2 Division of Human Nutrition and Epidemiology, Wageningen Agricultural University, The Netherlands; and 2The Wageningen Centre for Food Sciences

1

Objective: To investigate whether intake of Lactobacillus acidophilus strain L-1 lowers serum cholesterol in healthy men and women. Design: Randomised, placebo-controlled parallel trial. Setting: Subjects were free-living. Blood sampling and distribution of yoghurts were administered at a local hospital. Subjects: Seventy-eight adult men and women with cholesterol levels of 3.9 ± 7.8 mmol=L (mean s.d., 5.4 0.7). Interventions: Subjects consumed 500 mL of control yoghurt daily for two weeks. They were then randomly allocated to receive 500 mL per day of control yoghurt or of yoghurt enriched with Lactobacillus acidophilus L-1 for another six weeks. The yoghurts were spiked with a trace of lithium; compliance as assessed by plasma lithium was excellent. Results: Energy and nutrient intake was constant, and identical for the two groups. Mean body weight was stable. Baseline blood lipid concentrations in the control and treatment groups were highly similar. The effect of consumption of Lactobacillus acidophilus L-1 vs control on total cholesterol was 70.02 mmol=L (95% CI, 70.18 ± 0.15) after three weeks and 0.04 mmol=L (95% CI, 70.12 ± 0.20) after six weeks. Serum LDL and HDLcholesterol and triacylglycerol levels were also unaffected. Conclusions: Yoghurt enriched with Lactobacillus acidophilus L-1 does not lower serum cholesterol in men and women with normal to borderline high cholesterol levels. Sponsorship: Dutch Dairy Foundation on Nutrition and Health. Descriptors: lactobacillus acidophilus; blood lipids; yoghurt

Introduction The potential bene®cial effects of the so-called `probiotic' bacteria has raised much interest (Fuller, 1989; Orrhage et al, 1995; Saxelin, 1997) with cholesterol-lowering as a focus of research (Agerbñk et al, 1995; Richelsen et al, 1996; Lin et al, 1989; Schaafsma et al, 1998; Mital & Garg, 1995). Commercial products with Lactobacillus acidophilus that claim to lower serum cholesterol are marketed by several major dairy companies in Europe (Anonymous, 1996). However, this claim is mainly based on in vitro and animal studies. Several Lactobacillus acidophilus strains of human origin are able to remove cholesterol from culture media during growth in the presence of bile (Buck & Gilliland, 1994; Gilliland et al, 1985; Walker & Gilliland, 1993). The mechanism of action is probably that the bacteria deconjugate bile acids which, when deconjugated, coprecipitate with cholesterol at a pH < 5.5 (Gilliland et al, 1985; Klaver & Van der Meer, 1993; Tahri et al, 1995). If this also occurs in vivo, such coprecipitation might increase faecal excretion of bile acids and cholesterol. It is thought that the liver will compensate for the loss of bile acids by converting cholesterol into new bile acids. This *Correspondence: Dr NM de Roos, Wageningen Agricultural University, Division of Human Nutrition and Epidemiology, Bomenweg 2, 6703 HD Wageningen, The Netherlands. Received 7 August 1998; revised 13 November 1998; accepted 16 November 1998

conversion of cholesterol into bile acids might lower serum cholesterol levels. Evidence for a serum cholesterol lowering effect of Lactobacillus acidophilus in humans is sparse. Lin et al, 1989 found a cholesterol lowering effect of a mixture of Lactobacillus acidophilus and Lactobacillus bulgaricus in a 16-week parallel, placebo-controlled pilot trial with 38 subjects, but no effect in a subsequent 15-week double blind, cross-over trial with 344 subjects (Lin et al, 1989). Consumption of yoghurt with Lactobacillus acidophilus in a cross-over study with 30 healthy men resulted in 4.4% lower mean cholesterol concentrations than consumption of a placebo (Schaafsma et al, 1998). We investigated whether Lactobacillus acidophilus lowered serum cholesterol in a placebo-controlled, doubleblind study with healthy volunteers. We chose volunteers with normal to borderline-high serum cholesterol levels because this is the target group for these yoghurts and because most dietary components that can lower serum cholesterol do so also in normocholesterolaemic subjects (Lu et al, 1997; Pelletier et al, 1995). We chose a treatment period of six weeks because the same period or shorter was chosen in previous studies (Agerbñk et al, 1995; Lin et al, 1989; Schaafsma et al, 1998). To our knowledge every dietary component that lowers cholesterol does so within six weeks. We chose the maximum dose that could be given in a yoghurt.

Lactobacillus acidophilus L-1 does not lower blood lipids NM de Roos et al

278

Subjects and methods Prior approval for this study was obtained from the Medical Ethical Committee of the Department of Human Nutrition, Wageningen Agricultural University. All volunteers gave their written informed consent. Volunteers who completed the study received a ®nancial reward. Subjects Eligible for the study were healthy men and women between 18 and 65 y old who were free of heart disease, diabetes, liver or kidney disease, and did not use any medications known to affect blood lipid metabolism. Ninety-nine volunteers were screened and 85 were enrolled. Volunteers with a serum total cholesterol concentration higher than 8 mmol=L or a triacylglycerol concentration higher than 4 mmol=L were excluded. We made sure that at least 50% of the enrolled volunteers had serum cholesterol levels over 5 mmol=L, the cut-off value for mild hypercholesterolaemia in the Netherlands (Anonymous, 1992). Three subjects dropped out because of vacations, two because of illness, one because she could not eat all the yoghurt and one because of gastrointestinal complaints. The study was completed by 39 subjects in each of the two test groups. Study design After a two-week run-in period, in which the subjects added 500 mL of control yoghurt to their daily diet, subjects were strati®ed according to sex, age and LDL-cholesterol. Subjects in each stratum were randomly allocated to receive either control or Lactobacillus acidophilus yoghurt for six weeks. Venous blood samples were taken after an overnight fast at the end of the run-in period and at three and six weeks after the start of the test-period. Blood samples were taken twice in each week at an interval of 2 ± 4 d, which amounted to a total of six blood samples during the study. Yoghurts The control yoghurt was a non-fat yoghurt with a starter culture of Streptococcus thermophilus. The test yoghurt was control yoghurt enriched with Lactobacillus acidophilus L-1. The yoghurt was made once a week (NIZO, the Netherlands) and was given fresh, not pasteurised. Subjects were instructed to take 200 mL of the yoghurt directly after breakfast and 300 mL directly after dinner. Samples of both yoghurts were analysed on Rogosa medium every week. Levels of Lactobacillus acidophilus ranged from 4.8 6 109 to 2.7 6 1010 colony forming units per 500 mL. Measurements Blood samples were allowed to clot at room temperature and were centrifuged for 10 min at 1500 6 g. Serum samples were stored at 780 C. Serum lipids were measured or computed as described previously (Siedel et al, 1983; Warnick et al, 1982; Sullivan et al, 1985), with an intra-run coef®cient of variance of 2.6% and an inter-run coef®cient of variance of 3.8%. Body weight was measured with subjects clothed but without shoes. We used a single 24 h recall method to estimate energy and macronutrient intake in all subjects.

Compliance Lithium chloride (250 mmol) was added to the yoghurt (500 mL) as a marker (Sanchez-Castillo et al, 1987). Toxic effects of lithium below the therapeutic dose have not been reported (Amdisen, 1977; Lehmann, 1994). The two serum samples of week three were pooled for lithium analysis, and so were the two samples of week six. Serum samples were deproteinised by mixing ®ve parts with one part of 20% (w=v) trichloroacetic acid. Lithium concentrations in deproteinised serum samples were measured by atomic absorption spectrophotometry (Miller et al, 1989). Subjects recorded missed supplements, illness, and use of medications in a diary. Statistical analyses The two-sample t-test was used to compare changes in the acidophilus group with changes in the control group.

Results Each of the two test groups consisted of 11 men and 28 women with a mean age of 40 y and on average normal body weights and serum cholesterol values (Table 1). Blood lipid concentrations remained stable in the control group and were unaffected by consumption of Lactobacillus acidophilus L-1 (Table 2). The net effect of consumption of Lactobacillus acidophilus L-1, computed as the difference in change in blood lipid concentrations between acidophilus group and control group, was close to zero (Figure 1). We did not perform separate analyses for men and women because our study was not designed to do this. Mean changes in blood lipids, however, were almost identical in men and women. Twenty-one subjects in the control group and 21 subjects in the acidophilus group had initial serum cholesterol levels greater than 5.2 mmol=L (200 mg=dL, mean 5.7 0.4 mmol=L), the cut-off value for borderline high blood cholesterol in the USA (Anonymous, 1993). Lactobacillus acidophilus had no effect on serum cholesterol levels in this subgroup either: the net effect on total cholesterol was an elevation of 0.01 mmol=L (95% CI, 70.19± 0.21) after three weeks and 0.08 mmol=L (95% CI, 70.12± 0.29) after six weeks. Effects on other blood lipids were even smaller (data not shown). Changes in body weight were negligible; the mean difference in weight change between the acidophilus and control group was 0.04 kg (95% CI, 70.37± 0.44) at week three and 0.21 kg (95% CI, 70.35± 0.78) at week six.

Table 1 Mean ( s.d.) baseline characteristics of 78 subjects who completed a six-week study into the effect of consumption of Lactobacillus acidophilus L-1 on serum cholesterol concentrations Treatment Characteristic Men=women (number) Age (y) Body mass index (kg=m2) Total cholesterol (mmol=L)

Control

Lactobacillus acidophilus L-1

11=28 39.9 8.7 24.2 3.0 5.4 0.7

11=28 39.8 8.4 23.9 2.6 5.3 0.8

Conversion factors: cholesterol, 5mmol=L 193mg=dL.


Table 2 Blood lipid concentrations after the two-week run-in period and after three and six weeks of consumption of control yoghurt or yoghurt enriched with Lactobacillus acidophilus L-1 Treatment Control Sample Total cholesterol HDL-cholesterol LDL-cholesterol Triacylglycerol

Run-in Week 3 Week 6 Run-in Week 3 Week 6 Run-in Week 3 Week 6 Baseline Week 3 Week 6

Lactobacillus acidophilus L-1 (mmol=L)

5.19 0.74 5.27 0.74 5.12 0.69 1.51 0.40 1.53 0.44 1.54 0.41 3.12 0.80 3.17 0.82 3.04 0.80 1.22 0.52 1.22 0.51 1.17 0.54

5.12 0.72 5.19 0.74 5.10 0.74 1.40 0.32 1.43 0.35 1.40 0.35 3.20 0.68 3.26 0.68 3.18 0.69 1.13 0.66 1.10 0.57 1.11 0.54

Values are means s.d. of 39 subjects in the control group and 39 subjects in the Lactobacillus acidophilus L-1 group. Conversion factors: Cholesterol, 5mmol=L 193mg=dL; Triacylglycerol 1mmol=L 88.6mg=dL.

Mean intake of total energy, macronutrients, ®bre and cholesterol as measured by 24 h recall was not different between the test groups. Compliance, de®ned as a rise in serum lithium of at least three times the baseline value, was excellent. Mean serum lithium levels at three and six weeks were about 6.5-fold higher than baseline values. Two subjects from the acidophilus group had an increase from baseline of less than a factor three: one of them at week three and six, the other only at week six. Excluding these subjects from the analysis did not affect the outcome of the study. Discussion Lactobacillus acidophilus is the best studied of the `probiotic' cholesterol-lowering bacteria (Lin et al, 1989; Schaafsma et al, 1998; Harrison & Peat, 1975; Khedkar et al, 1993; de Rodas et al, 1996). Of all the Lactobacillus acidophilus-strains the L-1 strain, used in commercial products in the Netherlands, is thought to be one of the most effective strains to lower serum cholesterol

Figure 1 Net effects (means and 95% con®dence intervals) on blood lipid concentrations of daily consumption of yoghurt enriched with Lactobacillus acidophilus L-1 for six weeks. Net effects were computed as changes in blood lipid concentrations in the acidophilus group (39 subjects) minus changes in the control group (39 subjects).

(Buck & Gilliland, 1994). Our study, however, shows that daily consumption of yoghurt enriched with a high dose of Lactobacillus acidophilus L-1 for a period of six weeks does not lower blood lipid concentrations in men and women with normal to borderline high serum cholesterol levels. The 95% con®dence intervals for the net effect of the acidophilus yoghurt on serum LDL levels make any change of more than 0.15 mmol=L (4.7%) unlikely. The compliance of the subjects, checked by addition of a trace of lithium to the yoghurts, was excellent. Weight changes were negligible and background food intake was the same for the two groups. Choice of subjects with normal to borderline high serum cholesterol levels We do not think that the absence of an effect of the Lactobacillus acidophilus in our study was due to low initial cholesterol levels in our subjects. Feeding studies with fatty acids (Lu et al, 1997; Perez-Jimenez et al, 1995; Abbey et al, 1994; Cobb & Risch, 1993; Chan et al, 1991), fat restriction (Lefevre et al, 1997; Jenkins et al, 1997; Miller et al, 1998; Schaefer et al, 1995), psyllium (Ganji & Kies, 1996), or phytosterols (Pelletier et al, 1995), have shown that serum total cholesterol can be reduced in subjects with comparable or even lower mean serum cholesterol levels. Also, the subgroup with a mean total cholesterol level of 5.7 mmol=L (220 mg=dL) did not respond to Lactobacillus acidophilus either. It is unlikely that patients with borderline high blood cholesterol will bene®t from consumption of Lactobacillus acidophilus. Choice and dosage of probiotic strain It is possible that the strain we used in our study is not the most potent strain for cholesterol reduction. However, the strain used in this present study had the highest ability to remove cholesterol from a medium in vitro (Buck & Gilliland, 1994; Gilliland et al, 1985; Walker & Gilliland, 1993). The dose we gave, 109 ± 1010 colony-forming units per day, was up to 1000 times higher than that of previous studies aimed at cholesterol-lowering (Agerbñk et al, 1995; Lin et al, 1989; Schaafsma et al, 1998; Harrison & Peat, 1975; Khedkar et al, 1993). It is not possible to incorporate larger amounts into yoghurts or similar foods. Length of treatment period After a run-in period of two weeks we measured serum cholesterol after three and six weeks of treatment to measure short-term as well as long-term effects. Although we cannot exclude an effect of Lactobacillus acidophilus after a longer treatment period, we are not aware of any dietary component that has no effect for up to six weeks and then starts to lower serum cholesterol after more than six weeks. The hypothesis for the cholesterol lowering effect of Lactobacillus acidophilus L-1 is based on its ability to deconjugate bile acids in vitro. At a pH lower than 5.5 deconjugated bile acids and cholesterol coprecipitate and are excreted in faeces (Klaver & Van der Meer, 1993). However, the neutral to alkaline pH of the small intestine makes precipitation of cholesterol and deconjugated bile acids in vivo unlikely. Indeed, consumption of a fermented dairy product containing Lactobacillus acidophilus did not increase the relative amount of deconjugated bile salts in faeces of 8 ileostomy patients (Marteau et al, 1995).

279


280

Conclusions Yoghurt containing Lactobacillus acidophilus strain L-1 did not lower blood lipids in men and women with normal and borderline high blood cholesterol. Claims to such an effect would not be justi®ed. More in general, claims made by commercial producers that probiotic yoghurts lower serum cholesterol in man should not be based solely on in vitro and animal studies but should be supported by proper randomised trials in man. Acknowledgements ÐWe thank the Dutch Dairy Foundation on Nutrition and Health for funding this study. We are greatly indebted to our volunteers for taking part in this study and to staff and personnel of the Hofpoort Hospital in Woerden, the Netherlands, for screening the volunteers and for taking blood samples. We would like to thank Jeanne de Vries for advising us about the 24 h recalls. Finally we thank Robert Hovenier, Frans Schouten and Rijkelt Beumer for analysing blood and yoghurt samples.

References Abbey M, Noakes M, Belling GB & Nestel PJ (1994): Partial replacement of saturated fatty acids with almonds or walnuts lowers total plasma cholesterol and low-density-lipoprotein cholesterol. Am. J. Clin. Nutr. 59, 995 ± 999. Agerbñk M, Gerdes LU & Richelsen B (1995): Hypocholesterolaemic effect of a new fermented milk product in healthy middle-aged men. Eur. J. Clin. Nutr. 49, 346 ± 352. Amdisen A (1977): Serum level monitoring and clinical pharmacokinetics of lithium. Clin. Pharmacokinet. 2, 73 ± 92. Anonymous (1992): Herziening Consensus Cholesterol (Revised cholesterol consensus). Hart Bull. 23, 9 ± 21. Anonymous (1993): Summary of the second report of the National Cholesterol Education Program (NCEP). Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel II). J. Am. Med. Ass. 269, 3015 ± 3023. Anonymous (1996): New product initiatives in functional and healthy foods. Int. Food Ingred. 5, 16. Buck LM & Gilliland SE (1994): Comparisons of freshly isolated strains of Lactobacillus acidophilus of human intestinal origin for ability to assimilate cholesterol during growth. J. Dairy Sci. 77, 2925 ± 2933. Chan JK, Bruce VM & McDonald BE (1991): Dietary alpha-linolenic acid is as effective as oleic acid and linoleic acid in lowering blood cholesterol in normolipidemic men [see comments]. Am. J. Clin. Nutr. 53, 1230 ± 1234. Cobb MM & Risch N (1993): Low-density lipoprotein cholesterol responsiveness to diet in normolipidemic subjects. Metabolism 42, 7 ± 13. de Rodas BZ, Gilliland SE & Maxwell CV (1996): Hypocholesterolemic action of Lactobacillus acidophilus ATCC 43121 and calcium in swine with hypercholesterolemia induced by diet. J. Dairy Sci. 79, 2121 ± 2128. Fuller R (1989): Probiotics in man and animals. A review. J. Appl. Bacteriol. 66, 365 ± 378. Ganji V & Kies CV (1996): Psyllium husk ®ber supplementation to the diets rich in soybean or coconut oil: hypocholesterolemic effect in healthy humans. Int. J. Food Sci. Nutr. 47, 103 ± 110. Gilliland SE, Nelson CR & Maxwell C (1985): Assimilation of cholesterol by Lactobacillus acidophilus. Appl. Env. Microb. 49, 377 ± 381. Harrison VC & Peat G (1975): Serum cholesterol and bowel ¯ora in the newborn. Am. J. Clin. Nutr. 28, 1351 ± 1355. Jenkins DJ, Popovich DG, Kendall CW, Vidgen E, Tariq N, Ransom TP, Wolever TM, Vuksan V, Mehling CC, Boctor DL, Bolognesi C, Huang J & Patten R (1997): Effect of a diet high in vegetables, fruit, and nuts on serum lipids. Metabolism 46, 530 ± 537. Khedkar CD, Garge RD, Mantri JM, Kulkarni SA & Khedkar GD (1993): Effect of feeding acidophilus milk on serum cholesterol in human volunteers. J. Dairying, Foods & Home Sci. 12, 33 ± 38. Klaver FAM & Van der Meer R (1993): The assumed assimilation of cholesterol by Lactobacilli and Bi®dobacterium bi®dum is due to their bile salt-deconjugating activity. Appl. Environ. Metab. 59, 1120 ± 1124.

Lefevre M, Ginsberg HN, Kris-Etherton PM, Elmer PJ, Stewart PW, Ershow A, Pearson TA, Roheim PS, Ramakrishnan R, Derr J, Gordon DJ & Reed R (1997): ApoE genotype does not predict lipid response to changes in dietary saturated fatty acids in a heterogeneous normolipidemic population. The DELTA Research Group. Dietary Effects on Lipoproteins and Thrombogenic Activity. Arterioscler. Thromb. Vasc. Biol. 17, 2914 ± 2923. Lehmann K (1994): Endogenous lithium levels. Pharmacopsychiatry 27, 130 ± 132. Lin SY, Ayres JW, Winkler W & Sandine WE (1989): Lactobacillus effects on cholesterol: In vitro and in vivo results. J. Dairy Res. 72, 2885 ± 2899. Lu Z, Hendrich S, Shen N, White PJ & Cook LR (1997): Low linolenate and commercial soyabean oils diminish serum HDL cholesterol in young free-living adult females. J. Am. Coll. Nutr. 16, 562 ± 569. Marteau P, Gerhardt MF, Myara A, Bouvier E, Trivin F & Rambaud J-C (1995): Metabolism of bile salts by alimentary bacteria during transit in the human small intestine. Microb. Ecol. Health Dis. 8, 151 ± 157. Miller M, Teter B, Dolinar C & Georgopoulos A (1998): An NCEP II diet reduces postprandial triacylglycerol in normocholesterolemic adults. J. Nutr. 128, 582 ± 586. Miller NL, Durr JA & Alfrey AC (1989): Measurement of endogenous lithium levels in serum and urine by electrothermal atomic absorption spectrometry: a method with potential clinical applications. Anal. Biochem. 182, 245 ± 249. Mital BK & Garg SK (1995): Anticarcinogenic, hypocholesterolemic, and antagonistic activities of Lactobacillus acidophilus. Crit. Rev. Microbiol. 21, 175 ± 214. Orrhage K, Lidbeck A & Rafter J (1995): In¯uence of the probiotics, lactobacilli and bi®dobacteria, on gastrointestinal disorders in adults. In: Role of Gut Bacteria in Human Toxicology and Pharmacology. MJ Hill (ed). Slough, UK: Taylor & Francis pp 263 ± 280. Pelletier X, Belbraouet S, Mirabel D, Mordret F, Perrin JL, Pages X & Debry G (1995): A diet moderately enriched in phytosterols lowers plasma cholesterol concentrations in normocholesterolemic humans. Ann. Nutr. Metab. 39, 291 ± 295. Perez-Jimenez F, Espino A, Lopez-Segura F, Blanco J, Ruiz-Gutierrez V, Prada JL, Lopez-Miranda J, Jimenez-Pereperez J & Ordovas JM (1995): Lipoprotein concentrations in normolipidemic males consuming oleic acid-rich diets from two different sources: olive oil and oleic acid-rich sun¯ower oil. Am. J. Clin. Nutr. 62, 769 ± 775. Richelsen B, Kristensen K & Pedersen SB (1996): Long-term (6 months) effect of a new fermented milk product on the level of plasma lipoproteins Ð a placebo-controlled and double blind study. Eur. J. Clin. Nutr. 50, 811 ± 815. Sanchez-Castillo CP, Seidell J & James WP (1987): The potential use of lithium as a marker for the assessment of the sources of dietary salt: cooking studies and physiological experiments in men. Clin. Sci. 72, 81 ± 86. Saxelin M (1997): Lactobacillus GG Ð A human probiotic strain with thorough clinical documentation. Food Rev. Int. 13, 293 ± 313. Schaafsma G, Meuling WJA, Van Dokkum W & Bouley C (1998): Effects of a milk product, fermented by Lactobacillus acidophilus and with fructo-oligosaccharides added, on blood lipids in male volunteers. Eur. J. Clin. Nutr. 52, 436 ± 440. Schaefer EJ, Lichtenstein AH, Lamon-Fava S, Contois JH, Li Z, Rasmussen H, McNamara JR & Ordovas JM (1995): Ef®cacy of a National Cholesterol Education Program Step 2 diet in normolipidemic and hypercholesterolemic middle-aged and elderly men and women. Arterioscler. Thromb. Vasc. Biol. 15, 1079 ± 1085. Siedel J, Hagele EO, Ziegenhorn J & Wahlefeld AW (1983): Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic ef®ciency. Clin. Chem. 29, 1075 ± 1080. Sullivan DR, Kruijswijk Z, West CE, Kohlmeier M & Katan MB (1985): Determination of serum triglycerides by an accurate enzymatic method not affected by free glycerol. Clin. Chem. 31, 1227 ± 1228. Tahri K, Ballongue J & Schneider F (1995): Effects of three strains of bi®dobacteria on cholesterol. Lett. Appl. Microbiol. 21, 149 ± 151. Walker DK & Gilliland SE (1993): Relationships among bile tolerance, bile salt deconjugation, and assimilation of cholesterol by Lactobacillus acidophilus. J. Dairy Sci. 76, 956 ± 961. Warnick GR, Benderson J & Albers JJ (1982): Dextran sulfate-Mg2 precipitation procedure for quantitation of high-density-lipoprotein cholesterol. Clin. Chem. 28, 1379 ± 1388.