http://informahealthcare.com/ijfISSN: 0963-7486 (print), 1465-3478 (electronic)

Int J Food Sci Nutr, Early Online: 1–4! 2014 Informa UK Ltd. DOI: 10.3109/09637486.2014.945155

BRIEF COMMUNICATIONS

Effect of traditional leafy vegetables on the growth of lactobacilli andbifidobacteria

Muhammad Arshad Kassim1,2, Himansu Baijnath1, and Bharti Odhav1

1Department of Biotechnology and Food Technology, Durban University of Technology, Durban, South Africa and 2School of Molecular and Cell

Biology, University of the Witwatersrand, Johannesburg, South Africa

Abstract

Traditional leafy vegetables, apart from being a staple in the diet of most of sub-Saharan Africa,are an essential part of traditional medicine and are used daily by traditional healers in theregion to treat a wide variety of ailments. In this study, a batch culture technique was usedto investigate whether 25 infusions from 22 traditional leafy vegetables stimulated the growthof Lactobacillus bulgaricus, Lactobacillus lactis, Lactobacillus reuteri and Bifidobacterium longumin pure culture. High performance liquid chromatography was used to determine the inulincontent of the infusions. Sonchus oleraceus stimulated all four strains and Taraxacum officinalestimulated three strains. In total, 18 plants stimulated at least one of the four probiotic strains.The inulin content of the infusions varied between 2.5% and 3.6%, with Asparagus sprengericontaining the highest percentage. These results indicate that traditional leafy vegetablesdo stimulate the growth of the selected lactobacilli and bifidobacteria in pure culture andcontain inulin. These infusions can now be tested for prebiotic potential using mixed culturesystems or human hosts.

Keywords

Bifidobacteria, inulin, lactobacilli, Sonchusoleraceus, Taraxacum officinale

History

Received 31 March 2014Revised 12 June 2014Accepted 4 July 2014Published online 1 August 2014

Introduction

Traditional leafy vegetables, also known as indigenous orunconventional leafy vegetables, are a staple in the diet of thepoor in rural sub-Saharan Africa. These plants, often consideredby people from developed nations as weeds, are to the poor,their only source of micronutrients necessary for maintaininga balanced diet (Moyo et al., 2013; Uusiku et al., 2010).These vegetables are easy to grow – often found growingabundantly in the wild and do not require much water or carewhen cultivated.

Consumption of these plants has been found to have severalbenefits. However, studies thus far have concentrated only ontheir anti-oxidant, anti-bacterial, anti-fungal and nutritionalproperties (Akula & Odhav, 2008; Mensah et al., 2008; Odhavet al., 2007). No work has been done on their potential prebioticproperties.

The current research aimed to screen infusions from selectedtraditional leafy vegetables to identify if they stimulated thegrowth of selected lactobacilli and bifidobacteria in pure cultureand to investigate the inulin content of the plants as a potentialcontributor toward any stimulatory effect. The most promisingtraditional leafy vegetables can be further tested for their prebioticpotential in the future.

Materials and methods

Probiotic strains used

Lactobacillus lactis (Durban University of Technology culturecollection), Lactobacillus bulgaricus (Durban University ofTechnology culture collection) and Lactobacillus reuteri (ATCC55730) were grown aerobically at 37 �C while Bifidobacteriumlongum (Rosell-175 ME) was grown anaerobically at 37 �C in ananaerobic jar (Anaerocult� A, Merck, Darmstadt, Germany).

Collection and preparation of plant material

Twenty-two traditional leafy vegetables were identified by Prof.H. Baijnath using taxonomic keys and collected from Kwa-ZuluNatal, South Africa (Table 1). The rationale for selecting theseleafy vegetables is that they are regularly consumed by the poorin rural areas.

Fresh healthy looking leaves were harvested from all the plantsexcept for A. sprengeri, from which only tubers were harvested.Roots and leaves were harvested from S. oleraceus and T.officinale, and leaves and bulbs were harvested from T. violacea.The harvested material was placed in trays in a 30 �C oven withbuilt-in extractor until they had dried. The dried material was thenmilled using a blender (Salton) to obtain a powder-like texture andstored in glass bottles in a dark cupboard.

Preparation of the traditional leafy vegetable infusions

This was carried out according to the method outlined byTrojanova et al. (2004). Dried ground plant material (1 g) wasinfused in 5 ml of distilled water (85 �C) for 20 min. The infusionwas then filtered with a Whatman no. 1 filter paper. For high

Correspondence: Muhammad Arshad Kassim, School of Molecular andCell Biology, University of the Witwatersrand, Johannesburg, Private Bag3, Wits 2050, South Africa. Tel: +2711 717 6369. Fax: +2711 717 6351.E-mail: muhammad.kassim@wits.ac.za

Int J

Foo

d Sc

i Nut

r D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Let

hbri

dge

on 0

8/17

/14

For

pers

onal

use

onl

y.

performance liquid chromatography (HPLC), the infusionwas cooled to 60 �C and used immediately while for batchculture experiments, the infusion was cooled to room temperaturebefore use.

Evaluation of the effect of the infusions on the growth ofselected probiotics using a batch culture technique

The effect of the infusions on the growth of selected probioticsin pure culture was evaluated according to the method outlinedby Molina et al. (2005). MRS broth (Biolab; Merck, Gauteng)(5 ml) was inoculated with five loops-full of either L. bulgaricus,L. reuteri, L. lactis or B. longum and incubated overnight at 37 �Cunder aerobic conditions for lactobacilli and anaerobic conditionsfor bifidobacteria.

The overnight culture was standardised to 0.8 nm at 620 nmusing a Varian-Carey UV–Vis spectrophotometer by dilutingthe culture (if necessary) with sterile MRS broth until the desiredabsorbance (sufficient to give a concentration of 105 cfu/ml at thebeginning of the experiment) was obtained.

MRS broth (195 ml) was prepared in a flask. Five millilitresof the infusion was added to the MRS broth and mixedthoroughly. A positive control made up of 1 g Chicory inulin(Sigma, Munich, Germany) dissolved in 5 ml of distilled water,and a negative control with 5 ml MRS broth was also used.Twenty five millilitres of each was transferred to five sterile100 ml Schott bottles. Each bottle was then inoculated with 250mlof the test organism, capped, mixed and incubated aerobically forlactobacilli and anaerobically for bifidobacteria at 37 �C.

The effect of the infusions on the growth of the selectedprobiotics was evaluated at 0, 24, 48, 72 and 96 h by removingone bottle and testing 1 ml for total number of bacteria. Bacteriawere enumerated by serially diluting to 10�7 (in triplicate) onRogosa agar plates using the spread plate technique. The plateswere then incubated at 37 �C for 48 h and the colony forming unitswere counted using a colony counter.

Plates containing between 30 and 300 cfu/ml were counted.The average of three replicates and their standard deviationwas then calculated. Those plates with less than 30 cfu/ml wereconsidered too few to count while those with more than 300 cfu/ml were considered too numerous to count.

Quantification of inulin using HPLC

The inulin content of the infusions was determined using themethod outlined by Vendrell-Pascuas et al. (2000), being the mostquantitative method for the measurement of inulin-type fructans.This method involved inulinase hydrolysis of all the fructanmaterials present in the infusion to glucose and fructose, followedby the measurement of these sugars by HPLC. The methodincorporated a sample blank (without inulinase hydrolysis) fromeach sample to subtract contributions of free fructose and fructosefrom sucrose.

The amount of inulin dietary fibre (IN) was calculated usingthe formula:

%IN ¼ ½A� ðF1 � F2Þ=P� � 100

for a sample containing no sucrose and

%IN ¼ ½A� ðF1 � F2 � F3Þ=P� � 100

for a sample containing sucrose, with F3¼ S/B, where:F1 is concentration of the total fructose (g/l);F2 is concentration of the free fructose (g/l);F3 is concentration of fructose from sucrose (g/l);S is concentration of sucrose (g/l);P is mean mass (g/l) of the test samples;A¼ 1.03 (empirical conversion factor for fructose to inulin,obtained from different dilutions of inulin hydrolysed withFructozyme enzyme and quantified using rhamnose as internalstandard);B¼ 2.13 (empirical conversion factor for fructose to sucrose,obtained from different dilutions of sucrose hydrolysed withinulinase enzyme and quantified using rhamnose as internalstandard).

Calculation and expression of results

The growth response obtained over the 96 h period was calculatedusing the area under the curve (AUC). Plants that performedbetter than the median for the negative control were regarded ashaving a stimulatory effect.

The AUC was calculated for all the samples using theTrapezoidal rule. The data were given as average colony forming

Table 1. Details of plants analysed.

Scientific name Family English name Part used

Solanum nigrum L. Solanaceae Common nightshade LeafPhysalis viscosa L. Solanaceae Grapeground cherry LeafMomordica balsamina L. Cucurbitaceae Balsam apple LeafAmaranthus spinosus L. Amaranthaceae Prickly amaranth LeafAmaranthus hybridus L. Amaranthaceae Cockscomb LeafAmaranthus dubius Mart. Ex Thell. Amaranthaceae Wild spinach LeafAsystasia gangetica (L.) T.Anderson Acanthaceae Hunter’s spinach LeafJusticia flava (Forssk.) Vahl Acanthaceae Yellow justicia LeafEmex australis Steinh. Polygonaceae Cape spinach LeafOxygonum sinuatum (Hochst. & Steud ex Meisn.) Dammer Polygonaceae Double thorn LeafBidens pilosa L. Asteraceae Blackjack LeafGalinsoga parviflora Cav. Asteraceae Gallant soldier LeafPortulaca oleracea L. Portulacaceae Purslane LeafSenna occidentalis (L.) Link Fabaceae Coffee senna LeafChenopodium album L. Chenopodiaceae Fat hen LeafCeratotheca triloba (Bernh.) Hook.f. Pedaliaceae Wild foxglove LeafCentella asiatica (L.) Urb. Apiaceae Marsh pennywort LeafAsparagus sprengeri Regel Asparagaceae Asparagus fern TuberTulbaghia violacea Harv. Alliaceae Society garlic Leaf & BulbSonchus oleraceus (L.) L. Asteraceae Sowthistle Leaf & RootTaraxacum officinale (L.) Weber ex F.H. Wigg. Asteraceae Dandelion Leaf & RootCleome monophylla L. Capparidaceae Spindlepod Leaf

2 M. A. Kassim et al. Int J Food Sci Nutr, Early Online: 1–4

Int J

Foo

d Sc

i Nut

r D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Let

hbri

dge

on 0

8/17

/14

For

pers

onal

use

onl

y.

units at discrete time points from Time 0 h to 96 h. Each timesegment is considered a trapezoid and the area is given by thesegment width and the average number of colony forming unitswithin the segment width. The total AUC is the sum of the areasof the individual segments.

To confirm that the bacteria were comparable, the percentagechange from baseline in number of bacteria was calculated,and this percentage change was used in the calculation ofthe AUC.

Correlation analysis was used to describe the degree ofstrength by which inulin content of the infusions is linearly relatedto bacterial growth.

The Wilcoxon rank sum test was used to determine whetherthere was a significant difference (p50.05) between the medianAUC of the infusions and the medians of the positive and negativecontrols.

Results

A total of 21 infusions from 18 plants stimulated the growth of atleast one of the four probiotic strains. Lactobacillus bulgaricuswas stimulated by 19 of the 25 infusions, L. lactis by 15 andB. longum by two. L. reuteri was only tested against six infusionsbased on heuristic analysis, of which it was stimulated by five(Table 2).

Wilcoxon’s test revealed that the median growth responseof the four probiotic cultures combined was statistically signifi-cantly larger than the negative control for A. sprengeri tubers,S. oleraceus roots and T. officinale leaves and roots.

Inulin content was then measured in 17 of the infusions whichexhibited a stimulatory effect. The growth response (representedby AUC values) of all four probiotic strains was then combinedand the median thereof was correlated with the inulin contentof the infusions. A direct relationship (r¼ 0.1475) between thecombined growth response and the inulin content of the infusions

was observed. This was expected based on previous researchwhich concluded that inulin stimulates the growth of thesebacteria (Roberfroid et al., 2010) and hence inulin contributes tothe above observed stimulatory effect.

Discussion

The stimulatory effect of the infusions was more prominentamong the lactobacilli than the bifidobacteria tested in this study.Non-digestible carbohydrates of plant origin are the mainsubstrates of probiotic organisms and include resistant starch aswell as non-starch polysaccharides such as cellulose, hemicellu-lose, pectin and inulin (Blaut, 2002; Fooks & Gibson, 2002). Thediet of the poor in sub-Saharan Africa includes primarily leafyvegetables containing these non-digestible carbohydrates, whichare mainly fermented by lactobacilli and bifidobacteria (Mensahet al., 2008; Mussatto & Mancilha, 2007; Odhav et al., 2007),hence the above findings indicate that traditional leafy vegetablescould potentially contribute to the digestive health of thesepeople.

The inability of the majority of the inulin-containing infu-sions to stimulate the growth of B. longum may appearcontradictory to the accepted bifidogenic effect of inulin;however, this is most likely a function of the technique used.The fermentation of fructo-oligosaccharides in the colon is theresult of a complex sequence of metabolic pathways carried outby many bacterial species (Rossi et al., 2005). In this preliminarystudy however, each probiotic strain was tested individually asa pure culture. Further, more detailed research is necessaryand should involve the use of a mixed culture system orhuman host.

It is well established that inulin stimulates the growth oflactobacilli and bifidobacteria both in vitro and in vivo(Roberfroid, 2007). Inulin is also currently used as a prebioticon a commercial scale (Makelainen et al., 2010). However, plants

Table 2. Stimulatory effect and inulin content of traditional leafy vegetables.

Stimulated growth of*

Scientific name Plant part tested Inulin content (%) L. lactis L. bulgaricus L. reuteri B. longum

S. nigrum Leaf 2.8 3 3 not tested 5

P. viscosa Leaf 2.8 5 5 not tested 5

M. balsamina Leaf 2.8 5 3 not tested 5

A. spinosus Leaf 2.6 3 3 not tested 5

A. hybridus Leaf 2.8 3 3 not tested 5

A. dubius Leaf 2.7 5 3 not tested 5

J. flava Leaf 2.9 5 3 not tested 5

G. parviflora Leaf 2.8 5 3 not tested 5

C. triloba Leaf 3.1 3 3 not tested 5

A. sprengeri Tuber 3.6 3# 3 not tested 5

T. violacea Leaf 3.3 3# 3 5 5

Bulb 3.0 3 3 not tested 5

S. oleraceus Leaf 3.0 3# 3 3 3

Root 2.9 3 3 3 3

T. officinale Leaf 3.0 3# 3 3# 5

Root 2.9 3# 3 3# 5

C. monophylla Leaf 2.5 3 3 not tested 5

A. gangetica Leaf not tested 3 3 not tested 5

E. australis Leaf not tested 5 5 not tested 5

O. sinuatum Leaf not tested 3 5 not tested 5

B. pilosa Leaf not tested 5 5 not tested 5

P. oleracea Leaf not tested 5 5 not tested 5

S. occidentalis Leaf not tested 3 3 not tested 5

C. album Leaf not tested 5 5 3 5

C. asiatica Leaf not tested 5 3 not tested 5

*compiled using AUC values; 3 indicates bacterial growth greater than negative control; 5 indicates bacterial growth equal to or less than negativecontrol; #indicates bacterial growth equal to or greater than positive control.

DOI: 10.3109/09637486.2014.945155 Effect of traditional leafy vegetables on probiotics 3

Int J

Foo

d Sc

i Nut

r D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Let

hbri

dge

on 0

8/17

/14

For

pers

onal

use

onl

y.

contain many other compounds such as anti-oxidants andpolyphenols and recent findings suggest that anti-oxidants andpolyphenols may also elicit a stimulatory effect (Duda-Chodaket al., 2008; Hervert-Hernandez et al., 2009).

It is therefore possible that inulin was not solely responsiblefor the stimulatory effect observed in this study. The infusionsin this study were however only analysed for their inulin contentas it has been scientifically proven that inulin can stimulatethe growth of lactobacilli and bifidobacteria, and inulin hasalso been scientifically proven as a prebiotic (Roberfroid, 2007).For future work, a complete analysis of the infusion is recom-mended in order to provide a more detailed description of itscomposition.

Conclusion

This study has shown that selected traditional leafy vegetables,especially A. sprengeri, S. oleraceus, T. violacea, C. triloba andT. officinale, do contain inulin and can stimulate the growth of atleast one of the four probiotic strains in pure cultures. Four of theplants tested exhibited an effect higher than that of commerciallyavailable inulin. These infusions can now be tested for prebioticpotential by repeating these tests in vitro using a mixed culturesystem or in an animal or human host.

Further research will also be useful in the following areas:demonstrating the role of these plants in improving the quality ofprocessed foods and health; commercialisation; utilisation inunderdeveloped areas by local people as nutritional supplements;and educating communities on the benefits of consuming eitherthese plants as a food source or the inulin obtained from the plantsthat can be used as a functional food ingredient in order to attractthe worldwide market.

Acknowledgements

We would like to thank the National Research Foundation (NRF) andthe South Africa Netherlands research Programme on Alternativesin Development (SANPAD) for funding this project. We would liketo thank Dr. Viresh Mohanlall for his assistance with HPLC, and,Mr. Kevin Slaney and Mr. Ntutu Letseka for their assistance with thepreparation of this article.

Declaration of interest

This project was funded by National Research Foundation (NRF) andthe South Africa Netherlands research Programme on Alternatives inDevelopment (SANPAD). The authors declare that they have no conflictof interest.

References

Akula US, Odhav B. 2008. In vitro 5-lipoxygenase inhibition ofpolyphenolic antioxidants from undomesticated plants of SouthAfrica. J Med Plants Res 2:207–212.

Blaut M. 2002. Relationship of prebiotics and food to intestinalmicroflora. Eur J Nutr 41:11–16.

Duda-Chodak A, Tarko T, Statek M. 2008. The effect of antioxidants onLactobacillus casei cultures. Acta Sci Pol Technol Aliment 4:39–51.

Fooks LJ, Gibson GR. 2002. In vitro investigations of probiotics andprebiotics on selected human intestinal pathogens. FEMS MicrobiolEcol 39:67–75.

Hervert-Hernandez D, Pintado C, Rotger R, Goni I. 2009. Stimulatoryrole of grape pomace polyphenols on Lactobacillus acidophilusgrowth. Int J Food Microbiol 136:119–122.

Makelainen H, Saarinen M, Stowell J, Rautonen N, Ouwehand AC. 2010.Xylo-oligosaccharides and lactitol promote the growth ofBifidobacterium lactis and Lactobacillus species in pure cultures.Benef Microbe 2:139–148.

Mensah JK, Okoli RI, Ohaju-Obodo JO, Eifediyi K. 2008. Phytochemical,nutritional and medical properties of some leafy vegetables consumedby Edo people of Nigeria. Afr J Biotech 14:2304–2309.

Molina DL, Navarro-Martinez MD, Melgarejo FR, Hiner ANP,Chazarra S, Rodriguez-Lopez JN. 2005. Molecular properties andprebiotic effect of inulin obtained from artichoke (Cynara scolymus L).Phytochemistry 66:1476–1484.

Moyo M, Amoo SO, Ncube B, Ndhlala AR, Finnie JF, van Staden J. 2013.Phytochemical and antioxidant properties of unconventional leafyvegetables consumed in southern Africa. S Afr J Bot 84:65–71.

Mussatto S, Mancilha IM. 2007. Non-digestible oligosaccharides:a review. Carbohydr Polym 68:587–597.

Odhav B, Beekrum S, Akula US, Baijnath H. 2007. Preliminaryassessment of nutritional value of traditional leafy vegetables inKwa-Zulu Natal, South Africa. J Food Comp Anal 20:430–435.

Roberfroid M. 2007. Prebiotics: the concept revisited. J Nutr 137:830S–837S.

Roberfroid M, Gibson GR, Hoyles L, McCartney AL, Rastall R, RowlandI, Wolvers D, et al. 2010. Prebiotic effects: metabolic and healthbenefits. Brit J Nutr 104:1–63.

Rossi M, Corradini C, Amaretti A, Nicolini M, Pompei A, Zanoni S,Matteuzzi D. 2005. Fermentation of fructooligosaccharides and inulinby bifidobacteria:a comparative study of pure and fecal cultures. AppEnviron Microb 71:6150–6158.

Trojanova I, Rada V, Kokoska L, Vlkova E. 2004. The bifidogenic effectof Taraxacum officinale root. Fitoterapia 75:760–763.

Uusiku NP, Oelofse A, Duodu KG, Bester MJ, Faber M. 2010. Nutritionalvalue of leafy vegetables in sub-Saharan Africa and their potentialcontribution to human health. A review. J Food Comp Anal 23:499–509.

Vendrell-Pascuas S, Castellote-Bargallo AI, Lopez-Sabater MC. 2000.Determination of inulin in meat products by high performance liquidchromatography with refractive index detection. J Chromatogr 881:591–597.

4 M. A. Kassim et al. Int J Food Sci Nutr, Early Online: 1–4

Int J

Foo

d Sc

i Nut

r D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Let

hbri

dge

on 0

8/17

/14

For

pers

onal

use

onl

y.