Antihypertensive Effect of Lactotripeptides VPP and IPP in Randomized Placebo-Controlled Trials: A Comprehensive Meta-Analysis

GUANG Cuie1, SHANG Jiangang1, JIANG Bo1, Franco MILANI2, Robert Dick PHILLIPS3, ZHANG Hailing1

(1. State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi 214122, China; 2. Department of Food Science, University of Wisconsin-Madison, Madison 53706-1565, USA; 3. Department of Food Science and Technology,
University of Georgia, Griffin 30223-1797, USA)

Abstract: A comprehensive meta-analysis was conducted to evaluate the effect of lactotripeptides (LTPs) Val-Pro-Pro (VPP) and Ile-Pro-Pro (IPP) on blood pressure (BP) after oral administration. Twenty one randomized, placebo-controlled studies with 27 trials consisting of 2 142 prehypertensive or hypertensive individuals were identified through a defined search strategy, and due to heterogeneity among studies the random effect model was used for the analysis. The pool effect was a reduction of －1.78 mmHg [95% confidence interval (CI): －2.47 to －1.08, P < 0.001] for systolic blood pressure (SBP) and －0.67 mmHg (95% CI: －1.11 to －0.23, P < 0.001) for diastolic blood pressure (DBP), and no publication bias was present. Subgroup analyses showed that the intake of LTPs resulted in significant decreases in BP in Asians (SBP:
－6.56 mmHg; DBP: －3.01 mmHg), in subjects from publications before 2008 (SBP: －6.57 mmHg, DBP: －3.08 mmHg)
and participants with a dose of less than 5 mg/d (SBP: －5.20 mmHg; DBP: －2.05 mmHg), and in office SBP
(－2.74 mmHg) and DBP (－1.18 mmHg) and home SBP (－7.61 mmHg) and DBP (－4.83 mmHg), whereas no significant effects were observed in their respective counterparts. These results suggest that functional foods containing LTPs may be useful as part of dietary intervention of pre- or mild hypertension.

Key words: lactotripeptides; antihypertensive effect; functional foods; meta-analysis

基于综合性Meta分析的乳三肽VPP和IPP在随机安慰剂
对照实验中的抗高血压效能

光翠娥1，尚建钢1，江 波1，Franco MILANI2，Robert Dick PHILLIPS3，张海玲1

（1.江南大学 食品科学与技术国家重点实验室，江苏 无锡 214122；2.威斯康星大学食品科学系，
威斯康星 麦迪逊 53706-1565，美国；3.佐治亚大学食品科学与技术系，佐治亚 格里芬 30223-1797，美国）

摘 要：为评估口服乳三肽VPP（Val-Pro-Pro）和IPP（Ile-Pro-Pro）对血压的影响，基于综合性Meta分析选择了包含在21 个研究中的27 项临床实验、2 142 个高血压前期或高血压患者。由于存在异质性，采用随机影响模式进行分析。结果表明：Meta分析中未发现发表偏倚，收缩压（systolic blood pressure，SBP）降低了1.78 mmHg（95%可信区间为－2.47～－1.08，P＜0.001），舒张压（diastolic blood pressure，DBP）降低了0.67 mmHg（95%可信区间为－1.11～－0.23，P＜0.001）。亚组分析结果表明：亚洲人群SBP为－6.56 mmHg、DBP为
－3.01 mmHg，2008年前的研究对象SBP为－6.57 mmHg、DBP为－3.08 mmHg，口服乳三肽剂量＜5 mg/d
的患者SBP为－5.20 mmHg、DBP为－2.05 mmHg，以及诊所偶测SBP（－2.74 mmHg）与DBP（－1.18 mmHg）和家庭自测SBP（－7.61 mmHg）与DBP（－4.83 mmHg）都显著下降，而相应的比较对象则没有。因此，包含乳三肽VPP和IPP的功能性食品对高血压前期或轻度高血压患者是有效的，特别是在低剂量食用时对亚洲人群的诊所偶测血压和家庭自测血压有明显效果。

关键词：乳三肽；抗高血压效能；功能性食品；Meta分析

中图分类号：TS201.4 文献标志码：A 文章编号：1002-6630（2015）17-0212-07

doi:10.7506/spkx1002-6630-201517040

Hypertension is one of the most common diseases that afflict humans worldwide. It has numerous deleterious effects on human body, significantly increasing the risk of coronary artery disease, stroke, cardiac arrhythmia, heart failure, and abnormal renal function, and producing many other complications related to structural damage to the cardiovascular system[1]. Adoption of a healthy diet and lifestyle is important for the prevention of high blood pressure (BP) and an indispensable part of the management of hypertension.

Val-Pro-Pro (VPP) and Ile-Pro-Pro (IPP) were initially obtained by fermenting milk with Lactobacillus helveticus
(L. helveticus) and Saccharomyces cerevisiae (S. cerevisiae). VPP is located in the sequence f84−86 of β-casein and IPP in f74−76 of β-casein as well as f108−110 of κ-casein. Numerous in vivo studies have confirmed the antihypertensive effects of both VPP and IPP in spontaneously hypertensive rats and mildly hypertensive patients, and commercial products containing VPP and IPP are also available[2]. However, controversies in several meta-analyses occurred about their antihypertensive effects. Xu[3] and Jauhiainen[4] et al. confirmed a significant reduction of －4.0 −
－4.8 mmHg for systolic blood pressure (SBP) and about
－2.0 mmHg for diastolic blood pressure (DBP) but Pripp[5] did not; only slight significance (0.01 < P < 0.05) for SBP and DBP reductions occurred in the analysis of Cicero et al.[6]; Turpeinen et al.[7] only confirmed the BP-lowering effect in trials with small daily doses. Herein, a more comprehensive meta-analysis of randomized placebo-controlled trials was conducted to investigate the underlying antihypertensive effects of VPP and IPP.

1 Materials and Methods

1.1 Search strategy, eligible studies and data extraction

A systematic search of all English articles reporting the effects of VPP and IPP on BP in prehypertensive and stage 1 hypertensive subjects was conducted through the PubMed database, the internet and the reference lists. The terms used were as follows: BP, hypertension and milk or casein or peptide(s) or tripeptide(s) or IPP or VPP and intervention or randomized or trial. Studies that met the following criteria were eligible for a meta-analysis: 1) The study should be randomized, placebo-controlled; 2) the subjects should be defined as having high-normal BP (prehypertension) or hypertension; 3) the effects of VPP and IPP or food products containing them on BP should be reported; and 4) the intervention should last for at least 4 weeks. From each trial, author(s), number of participants, publication year, ethnicity, dosage, method of BP measurement, and baseline and end SBP and DBP for both active and placebo-controlled groups were extracted (Table 1). In the case that multiple methods of BP measurement were used, the results for the outcome determined as primary were selected.

1.2 Statistical analysis

The effect of intervention was estimated with an intra-subject comparison of participants’ BP at the start and end of trial. Statistically, let Xij be the BP for the intervention group and Yij for the placebo-controlled group, where i=1 or 2 which was at start or end of the trial, respectively, and j=1, 2,…, n was the participants with active intervention. The mean effect of intervention was defined as formulas (1) and (2).

(1)

and for placebo

(2)

For the meta-analysis, the trial effect could then be calculated by (

－

) with standard error of trial effect estimated by:

(3)

Where SDI and SDP were the standard deviations for the intervention and placebo-controlled groups, respectively[6].

The random effects weighted model by Dersimonian et al.[29] based on the inverse variance was used to calculate the pooled effect size with 95% confidence interval (CI). Heterogeneity among studies was evaluated with Cochran’s Q statistic based on chi-square distribution and the I2 parameter representing the percentage of total variation across studies that is attributable to heterogeneity rather than to chance. Sensitivity analysis was conducted by omitting one study at a time to see the extent to which inferences depend on a particular study or group of studies. Publication bias was visually examined after generation of a funnel plot, in which standard error was plotted against weighted mean difference (WMD) in BP, and further tested through Egger regression. The effect and quality of individual trials were examined by making cumulative forest plots. The meta-analysis was also repeated by dividing the available trials in terms of ethnicity of the enrolled patients (Asians vs. Caucasians), dose
(> 5 mg/d vs. < 5 mg/d), year published (before 2008 vs. since 2008) and the method of BP measurement (office, home and 24 h ambulatory BP). Statistical calculation for meta-analysis and construction of plots were conducted using the StataSE version 12.0 (Stata Corp, College Station, TX, USA), and weighted least square regression for meta-regression was performed using SPSS 13.0 for Windows (SPSS Inc., Chicago, IL, USA)[6].

2 Results and Analysis

2.1 Primary analysis

Fig.1 Standard forest plots from meta-analysis for the effects of VPP and IPP intake on systolic (a) and diastolic (b) blood pressures

Twenty one randomized, placebo-controlled studies with 27 trials that met our inclusion criteria were included in the primary analysis (Table 1). Among them, eighteen studies were conducted in a parallel study design, and three were cross-over studies; Nineteen studies were double-blind, and two were single-blind. These trials included
2 142 prehypertensive or hypertensive, mostly middle-aged individuals with a daily peptide dose and an intervention duration ranging from 2.0 to 52.5 mg (IPP+VPP) and from 4 to 21 weeks, respectively. Generally, VPP and IPP were incorporated into milk products, juice, tablets (in three studies) or spread (in one study).

Significant heterogeneity between different studies (Cochran’s Q test P < 0.001 and I2 > 56% for both SBP and DBP) was observed, and therefore a random effects model was used for the meta-analysis. Sensitivity analysis showed no substantial change of the pooled effect of LTPs on BP with an exclusion of any one of the studies. The forest plots of the primary analysis were shown in Fig.1 and the WMD for the supplementation of VPP and IPP on lowering BP was －1.78 mmHg (95% CI: －2.47 to －1.08, P < 0.001) for SBP, and －0.67 mmHg (95% CI: －1.11 to －0.23,
P < 0.001) for DBP, suggesting reductions of both SBP and DBP. The funnel plots of SBP and DBP (Fig.2) did not indicate a significant asymmetry, hence excluding the publication bias (Egger’s test P > 0.05 for both SBP and DBP).

WMD. weighted mean difference.

Fig.2 Funnel plots of mean effects on systolic (a) and diastolic (b) blood pressures to detect signs of publication bias

2.2 Subgroup analysis

Table 2 Results of meta-analysis for significance, heterogeneity and Egger’s tests

Tables 1 and 2 summarized the comparison of effect sizes in clinical subgroups. Of the 27 trials, thirteen were conducted in Japan and the remaining in Europe. It was discovered that the Asian group [SBP: －6.56 mmHg
(95% CI: －8.78 to －4.34), P < 0.001; DBP: －3.01 mmHg (95% CI: －4.01 to －2.01), P < 0.001] showed a much more obvious reduction in BP than the Caucasian one [SBP:
－0.18 mmHg (95% CI: －0.87 to 0.50), P = 0.422; DBP: －0.09 mmHg (95% CI: －0.55 to 0.37), P = 0.305] after the intake of LTPs (Fig.3), which is in accordance with the result for subjects in publications before 2008 and later. The WMD was －6.57 mmHg (95% CI: －8.66 to －4.47,
P < 0.001) for SBP and －3.08 mmHg (95% CI: －4.43 to －1.73, P < 0.001) for DBP for subjects in publications before 2008, whereas the WMD was －0.22 mmHg (95% CI: －0.90 to 0.47, P = 0.387) for SBP and －0.19 mmHg (95% CI:－0.66 to 0.27, P = 0.17) for DBP for those starting from 2008 (Fig.4). Also, the effect of LTPs on BP with a dose of lower than 5 mg/d [SBP: －5.20 mmHg
(95% CI: －6.85 to －3.56), P < 0.001; DBP: －2.05 mmHg
(95% CI: －3.02 to －1.08), P < 0.001] was more evident than with a dose of more than 5 mg/d [SBP: －0.48 mmHg (95% CI: －1.26 to 0.31), P = 0.145; DBP: －0.19 mmHg (95% CI: －0.67 to 0.29), P = 0.134] (Fig.5). Besides, the methodology employed for BP measurement influenced the BP lowering size of LTPs: for the home BP, significant reductions of －7.61 mmHg (95% CI:－13.56 to －1.67,
P < 0.05) for SBP and －4.83 mmHg (95% CI:－8.17 to
－1.48, P < 0.01) for DBP were observed; for the office BP, significant reductions of －2.74 mmHg (95% CI: －3.62 to －1.86, P < 0.001) for SBP and －1.18 mmHg (95% CI:
－1.75 to 0.60, P < 0.01) for DBP were also found; for the
24 h ambulatory BP measurement, no significant reduction was discovered for either SBP (－0.02 mmHg, 95% CI:
－1.41 to 1.37, P = 0.285) or DBP (0.23 mmHg, 95% CI: －0.70 to 1.16, P = 0.255) (Fig.6). The meta-regressions suggested that with exceptions for Asian DBP and home BP (Cochran’s Q test P > 0.05), significant heterogeneity existed for all the other factors across the studies (Table 2). Still, Egger’s tests did not indicate the existence of asymmetry for subgroups (P > 0.05 for both SBP and DBP) except for the home BP group with only two studies.

Fig.3 Standard forest plots from meta-analysis for the effects of VPP and IPP intake on systolic (a) and diastolic (b) blood pressures in Asian and Caucasian groups

Fig.4 Standard forest plots from meta-analysis for the effects of VPP and IPP intake on systolic (a) and diastolic (b) blood pressures in subjects from publications before 2008 and later

Fig.5 Standard forest plots from meta-analysis for the effects of VPP and IPP intake with different doses on systolic (a) and diastolic (b) blood pressures

Fig.6 Standard forest plots from meta-analysis for the effects of VPP and IPP intake on systolic (a) and diastolic (b) blood pressures measured by different methods

3 Conclusion and Discussion

Our meta-analysis of clinical trials showed that both VPP and IPP could lead to small but statistically significant reductions in SBP and DBP, which is more or less in line with the findings of Xu[3], Jauhiainen[4], Cicero[6] and Turpeinen[7] et al.. The less reduction in BP may be due to the presence of more trials with no significant effect of LTPs on BP in this analysis. Nevertheless, a decrease in SBP by about 2 mmHg is of clinical significance and would reduce the risk for stroke and myocardial infarction by about 4%[2]. Similarly, heterogeneity occurred, exclusion sensitivity analysis indicated no major impact of a specific trial on the pooled effect estimate and no sign of publication bias was present. However, in the meta-analysis of Pripp[5], the effect of daily intake of VPP and IPP was not significant and
P values of the Cochrane Q-test (0.16 and 0.08 for SBP and DBP, respectively) suggested no heterogeneity across studies with VPP and IPP. No heterogeneity occurred in the analysis of Xu et al.[3] and thus a fixed effects model was used.

The meta-regressions also suggested that ethnicity, publication year, dosage and BP measurement method could contribute to the heterogeneity and therefore were stratified. A difference in the racial response to LTPs, similar to the result of Cicero et al.[6], was observed. In our analysis, subjects in publications before 2008 showed a significant reduction in BP but those starting from 2008 did not, which further supports the difference of racial response since most of the trials for Caucasians with no significance in BP reduction have been reported since 2008. Our finding about the effect of publication date (before 2008) is in agreement with the BP lowering outcome of Xu[3] and Jauhiainen[4] et al. but is different from that of Pripp[5]. Trials included in these three analyses were all reported before 2008. The analyses of Xu[3] and Cicero[6] et al. only suggested no dose-dependency; our study found that a low dose of lower than 5 mg/d resulted in a significant reduction in BP after the intake of LTPs but a higher dose did not, indicating that no additional benefit on BP is gained with higher doses. This finding is also relevant for commercial food products due to the challenges that high doses pose on the sensory properties of the product as well as high price[7]. This was the first time to show the effect of methodology used for BP measurement on the BP lowering size after the oral administration of LTPs with significant reductions present in office SBP and home BP.

Compared to our study, the recent analysis by Turpeinen et al.[7] included 21 trials with only low dose (2.0−10.2 mg/d) and about 1 500 subjects; eighteen studies with 24 arms and
1 691 volunteers were involved in the analysis by Cicero et al.[6]; Jauhiainen et al.[4] included only 5 trials conducted by their own lab before 2008 in which 295 moderate hypertensives were involved; the analysis by Xu et al.[3] included 9 studies with 12 trials published between 1996 and 2005 and a total of 623 subjects; and Pripp[5] analyzed 15 clinical studies (published 1996-2007) with 826 participants, and VPP and IPP were investigated only in 9 of them. The main limitation of this meta-analysis is that it excluded the preliminary studies and the non-controlled trials with large sample size. For the included trials here, more details, if provided, such as BP data of the enrolled individuals, weight and diet patterns, could contribute to the final evaluation and explanation.

This meta-analysis shows that both VPP and IPP reduce blood pressure in prehypertensive and mildly hypertensive subjects, particularly office BP and home BP in Asians with a low dose. In combination with other lifestyle changes, products containing these tripeptides are thus useful as dietary treatment and prevention of hypertension.

References:

[1] Guang Cuie, PHILLIPS R D. Plant food-derived angiotensin Ⅰ converting enzyme inhibitory peptides[J]. Journal of Agricultural and Food Chemistry, 2009, 57: 5113-5120.

[2] Guang Cuie, Milani F, Shang Jiangang, et al. New evidences for production, biochemistry, in vitro and in vivo effects of tripeptides Val-Pro-Pro and Ile-Pro-Pro[J]. Journal of Food Agriculture and Environment, 2012, 10: 32-39.

[3] XU J, QIN L, WANG P, et al. Effects of milk tripeptides on blood pressure: a meta-analysis of randomized controlled trials[J]. Nutrition, 2008, 24: 933-940.

[4] Jauhiainen T, Korpela R, Vapaatalo H, et al. Bioactive milk peptides and blood pressure[J]. Agro FOOD Industry Hi Tech, 2009, 20: 26-28.

[5] Pripp A H. Effect of peptides derived from food proteins on blood pressure: a meta-analysis of randomized controlled trials[J]. Food and Nutrition Research, 2008, 52. doi: 10.3402/fnr.v52i0.1641.

[6] Cicero A F G, Gerocarni B, Laghi L, et al. Blood pressure lowering effect of lactotripeptides assumed as functional foods: a meta-analysis of current available clinical trials[J]. Journal of Human Hypertension, 2011, 25: 425-436.

[7] Turpeinen A M, Jarvenpaa S, Kautiainen H, et al. Antihypertensive effects of bioactive tripeptides: a random effects meta-analysis[J]. Annals of Medicine, 2013, 45: 51-56.

[8] Hata Y, Yamamoto M, Ohni M, et al. Placebo-controlled study of the effect of sour milk on blood pressure in hypertensive subjects[J]. The American Journal of Clinical Nutrition, 1996, 64: 767-771.

[9] Kajimoto O, Aihara K, Hirata H, et al. Hypotensive effects of the tablets containing “lactotripeptides (VPP, IPP)”[J]. Journal of Food and Nutrition Research, 2001, 4: 51-61.

[10] Kajimoto O, Nakamura Y, Yada H, et al. Hypotensive effects of sour milk in subjects with mild or moderate hypertension[J]. Journal of Japanese Society of Nutrition and Food Science, 2001, 54: 347-354.

[11] KAJIMOTO O, KUROSAKI T, MIZUTANI J, et al. Antihypertensive effects of liquid yogurts containing lactotripeptides (VPP, IPP) in mild hypertensive subjects[J]. Journal of Food and Nutrition Research, 2002, 5(3): 55-66.

[12] Hirata H, Nakamura Y, Yada H, et al. Clinical effect of new sour milk drink on mild or moderate hypertensive subjects[J]. Journal of New Remedies and Clinics, 2002, 51: 61-69.

[13] Seppo L, Kerojoki O, Suomalainen T, et al. The effect of a Lactobacillus helveticus LBK-16 H fermented milk on hypertension: a pilot study on humans[J]. Milchwissenschaft-Milk Science International, 2002, 57: 124-127.

[14] Seppo L, Jauhiainen T, Poussa T, et al. A fermented milk high in bioactive peptides has a blood pressure-lowering effect in hypertensive subjects[J]. American Journal of Clinical Nutrition, 2003, 77: 326-330.

[15] Mizushima S, Ohshige K, Watanabe J, et al. Randomized controlled trial of sour milk on blood pressure in borderline hypertensive men[J]. American Journal of Hypertension, 2004, 17: 701-706.

[16] Nakamura Y, Kajimoto O, Kaneko K, et al. Effects of the liquid yogurts containing “lactotripeptide (VPP, IPP)” on high normal blood pressure[J]. Journal of Food and Nutrition Research, 2004, 7: 123-127.

[17] Toumilehto J, Lindstrom J, Hyyrynen J, et al. Effect of ingesting sour milk fermented using Lactobacillus helveticus bacteria producing tripeptides on blood pressure in subjects with milk hypertension[J]. Journal of Human Hypertension, 2004, 18: 795-802.

[18] Aihara K, Kajimoto O, Hirata H, et al. Effect of powdered fermented milk with Lactobacillus helveticus on subjects with high-normal blood pressure or mild hypertension[J]. Journal of the American College of Nutrition, 2005, 24: 257-265.

[19] Jauhiainen T, Vapaatalo H, Poussa T, et al. Lactobacillus helveticus fermented milk lowers blood pressure in hypertensive subjects in 24-h ambulatory blood pressure measurement[J]. American Journal of Hypertension, 2005, 18: 1600-1605.

[20] Mizuno S, Matsuura K, Gotou T, et al. Antihypertensive effect of casein hydrolysate in a placebo-controlled study in subjects with high-normal blood pressure and mild hypertension[J]. British Journal of Nutrition, 2005, 94: 84-91.

[21] Sano J, Ohki K, Higuchi T, et al. Effect of casein hydrolysate, prepared with protease derived from Aspergillus oryzae, on subjects with high-normal blood pressure or mild hypertension[J]. Journal of Medicinal Food, 2005, 8: 423-430.

[22] Engberink M F, Schouten E G, Kok F J, et al. Lactotripeptides show no effect on human blood pressure: results from a double-blind randomized controlled trial[J]. Hypertension, 2008, 51: 399-405.

[23] van der Zander K, Bots M, Bak A A, et al. Enzymatically hydrolyzed lactotripeptides do not lower blood pressure in mildly hypertensive subjects[J]. The American Journal of Clinical Nutrition, 2008, 88: 1697-1702.

[24] van der Zander K, Jäkel M, Bianco V, et al. Fermented lac-totripeptides-containing milk lowers daytime blood pressure in high normal-to-mild hypertensive subjects[J]. Journal of Human Hypertension, 2008, 22: 804-806.

[25] van Mierlo L, Koning M, van der Zander K, et al. Lactotripeptides do not lower ambulatory blood pressure in untreated whites: results from 2 controlled multicenter crossover studies[J]. American Journal of Clinical Nutrition, 2009, 89: 617-623.

[26] de Leeuw P, van der Zander K, Kroon A, et al. Dose-dependent lowering of blood pressure by dairy peptides in mildly hypertensive subjects[J]. Blood Pressure, 2009, 18: 44-50.

[27] Yoshizawa M, Maeda S, Miyaki A, et al. Additive beneficial effects of lactotripeptides and aerobic exercise on arterial compliance in postmenopausal women[J]. American Journal of Physiology: Heart and Circulatory Physiology, 2009, 297: H1899-H1903.

[28] Cicero A F G, Rosticci M, Veronesi M, et al. Haemodynamic effects of lactotripeptides from fermentation by lactobacillus helveticus in normotensive subjects and patients with high-normal blood pressure: a randomized, double-blind, cross-over clinical trial[J]. Nutrition, Metabolism and Cardiovascular Diseases, 2009, 19. doi: 10.1016/S0939-4753(09)70026-2.

[29] DerSimonian R, Laird N. Meta-analysis in clinical trials[J]. Controlled Clinical Trials, 1986, 7: 177-188.