Int. J. Biosci.

2012

International Journal of Biosciences (IJB)

ISSN: 2220-6655 (Print) 2222-5234 (Online)
Vol. 2, No. 2, p. 45-52, 2012
http://www.innspub.net

RESEARCH PAPER OPEN ACCESS

Evaluation of the microbiological quality of the leaves of Solanum

macrocarpum L. cultivated with the chicken’s droppings and water of marsh
in Cotonou (Republic of Benin)
Victorien T. Dougnon1,3,*, Honoré S. Bankolé3, Patrick A. Edorh1,2, Jacques T. Dougnon4,
Modeste Gouissi1, Armelle Hounkpatin1, Sabine Montcho1, Hervé Azonhè5, Jean-Robert Klotoé3,
Michel Boko1
Interfaculty Center of Formation and Research in Environment for the Sustainable Development, Laboratory of Toxicology
1

and Environmental Health, University of Abomey-Calavi (UAC), 01 BP 1463 Cotonou, Benin

Faculty of Science and Technology, Department of Biochemistry and Cellular Biology, University of Abomey-Calavi (UAC),
2

01 BP 526 Cotonou, Benin

Polytechnic School of Abomey-Calavi, Department of Human Biology, University of Abomey-Calavi, 01 BP 2009 Cotonou,
3

Benin
Polytechnic School of Abomey-Calavi, Department of Animal Production and Health, Research Laboratory in Applied
4

Biology , University of Abomey-Calavi, 01 BP 2009 Cotonou, Benin

Faculty of Arts and Social Sciences, Department of Geography, University of Abomey-Calavi, Benin
5

Received: 29 December 2012

Revised: 02 February 2012
Accepted: 03 February 2012

Key words: Escherichia coli, salmonellas, coliforms, microbiology.

Abstract

This study has evaluated the microbiological quality of the leaves of Solanum macrocarpum L. cultivated with chicken’s droppings
and water of watering. A pilot site at Glo and market-gardening sites of Houeyiho, Fidjrosse and Agongbomey were used as sites of
study. The salmonellas and fecal coliforms whose Escherichia coli were required in the environment of culture of Solanum
macrocarpum L. Eight samples resulting respectively from water of watering; ground and leaves were analyzed. No salmonella was
detected on these samples. Water of watering of the pilot site of Glo is not contaminated by Escherichia coli compared to those of the
sites of Houeyiho (1.8.104 ± 28.42 UFC/100 ml) ; Fidjrosse (0.95.104 ±70.10 UFC/100 ml) ; Agongbomey (5545 ± 77.53 UFC/100 ml)
(p<0.05). The leaves of Solanum macrocarpum L. were contaminated by Escherichia coli with different degrees according to the
site: Glo (35.65.104 ± 457.08 UFC/g); Houeyiho (0,95.103 ± 70.71 UFC/g); Fidjrosse (1.103 ± 0 UFC/g) ; Agongbomey (0.9.103 ± 0
UFC/g) (p<0.05). In addition, Escherichia coli is strongly developed on the leaves of Solanum macrocarpum L. from the pilot site of
Glo with a neutral pH (7.305 ± 0.064) compared to the other sites: 6.550 at Houeyiho; 6.480 at Fidjrosse and 6.445 at Agongbomey
(p<0.05). The study has shown that the leaves of Solanum macrocarpum L. are contaminated by the enterobacteria and their
consumption may expose the populations to some risks of enteric diseases.
*Corresponding Author: Dougnon Tamègnon Victorien  [email protected]

Introduction

45 Dougnon et al.
 Int. J. Biosci. 2012

Urban agriculture became throughout the whole world Thus, inside the city, the population developed a local
and specifically in the developing countries, one of the strategy of market-gardening production which
activities necessary to ensure the food safety of the resulted in the use of the chicken’s droppings and
townsmen and to get financial resources with the water of the marshes to amend and sprinkle the
unemployed persons of the cities. Mougeot (2006) market-gardening products. These poultry’s droppings,
reported that between 2015 and 2020, more half of the as the many waste used in agriculture, contains organic
world population will live in urban or perish-urban matter and biogenic salts which constitute a
zone. The role of urban agriculture is fully justified contribution interesting for the grounds of culture
when taking into consideration this demographic (Legret et al., 1988). But the presence of pathogenic
explosion on a worldwide scale. Its contribution with in the droppings represents a major constraint with its
the food production on a world level which was at 15 % use in agriculture. Indeed, in the event of use of
in 1993 can have exceeded 33 % as from the year 2005 animals waste on the market gardenings (whose leaves
(Niang, 1996). are consumed), some pathogenic micro-organisms
could pass in the excreta of the people infected to find
Smit et al. (1996) revealed that this activity is practiced itself in water of the marshes being used for watering
in at least 90 towns of 31 countries of the Southeast of the cultures. The enterobacteria could be contained
Asia, of the Middle East, of Europa, of Sub-Saharan in these droppings and contaminate the plants which
Africa, Antilles and North, Center, South America. In they are used to amend (Métras, 2003). This situation
addition, 800 million people practice this activity on a could induce medical problems with the fresh
worldwide scale (Koc et al., 2006). The problems of vegetable consumers.
urban agriculture and in West Africa were the subject
of several publications which show well its importance Therefore, the conditions of production of the
in cities (Akinbamijo et al., 2002). Urban agriculture is vegetable-leaves on the market-gardening grounds at
thus socially very significant. The activities of Cotonou do not guarantee their sanitary quality. As
production, transformation and marketing offer remarked that more and more market-gardeners
employment opportunities for a significant mass of the prefer to use the droppings of chickens and water of
urban population in situation of chronic marsh to amend and sprinkle their cultures, the study
unemployment and the rural ones in seasonal evaluated that practice’s impact on the microbiological
migration. quality of Solanum macrocarpum L., a vegetable
highly appreciated at Cotonou (MVAD, 2003). It is
Benin, like many African countries, experienced these then about a contribution in favor of food safety in
last years a development of urban agriculture. In Republic of Benin.
Cotonou, the market-gardening production and more
precisely that of the vegetable-leaves is a reality Materials and methods
impossible to circumvent. It contributes to the Localization of the sites of study
provisioning of the markets of the city in food Our survey took in account a witness site at Glo (6° 56 '
products. The modernization and the intensification of Northern latitude and 2°30’ Eastern longitude) and
the systems of production induce a stronger use of market-gardening sites of Houeyiho (6° 21' 20''
manures and comparable. The producers answer this Northern latitude and 2° 21' 35'' Eastern longitude), of
request by the recycling of various types of waste Fidjrosse (6° 22' Northern latitude and 2° 24' Eastern
(Akinbamijo et al., 2002). longitude); of Agongbomey (6° 21' Northern latitude
and 2° 24' 45'' Eastern longitude) (Fig. 1).

46 Dougnon et al.
 Int. J. Biosci. 2012

samples of water, ground and leaves of Solanum

macrocarpum L. The fecal coliforms and especially
Escherichia coli were required because they are pilot
presence of pathogenic microorganisms.

Research of fecal coliforms and E.coli on the leaves of

Solanum macrocarpum L.
After having weighed 25 grams of the samples, we add

Fig. 1. Localization of sites included in the survey. Peptoned Plug Water (EPT) until the limit of 250
grams. That’s what we call pre-enrichment. Then we

Sampling carry out successive dilutions from 10 -2 to 10-4 starting

It has been used among others of the white sachets from pre-enrichment before sow by incorporation 1 ml

labeled to collect samples, of sterile gloves for of the dilutions of 10-3 and 10-4 on the medium Rapid

withdrawals, an icebox for the transportation of E.coli. It has been incubated with 44°C during 24

samples. For exams to the laboratory, we used hours ± 2 hours. All the colonies of E.coli appeared as

surroundings of culture as Rapid E.coli, some violet and those of the fecal coliforms were blue.

serological pipettes, drying ovens, the crushers.

On each site, we carried out in two different market- Research of fecal coliforms and E.coli on the samples

gardeners, some samples of ground; water of watering; of water of marsh

leaves of Solanum macrocarpum L. 500 grams of fresh It has been taken 1 ml of the sample and carried out

leaves were collected on the vegetable seedlings then successive dilutions with 10-1 and 10-2. Then it has been

introduced into hermetically tied plastic sachets. 500 taken 1ml of dilution with 10-2 and sow by

grams of ground were taken with a depth ranging incorporation on the medium Rapid E.coli. After

between 0 and 15 centimeters and put in plastic incubating with 44°C during 24 hours ± 2 hours, the

sachets. The samples of the water of watering were colonies of E.coli appeared as violets and those of the

directly taken in sterile sachets at a rate of 0.5 liter. All fecal coliforms were blue.

the samples were transported in a refrigerator towards

the laboratory in a one hour interval after the taking Research of fecal coliforms and E.coli on the samples

away where they were preserved at once at a of water of tap

temperature of 4° C. The microbiological analyses were We had to filter the sample and to deposit the filtrated

carried out in the 24 hours following the test sample membrane on the medium Rapid E.coli and incubate it

selections. The research for pH was made by the two with 44°C during 24 hours ± 2. The colonies of E.coli

weeks. appeared as violets and those of the fecal coliforms

were blue.

Methods of analysis
Microbiological analyses Research of salmonellas on the samples of leaves,

The microbiological analyses were carried out in the grounds and water

Hygiene Section of Water and Food at the Service of Either it is a sample of ground, water or leaf, the

the Biomedical Analyses Laboratories (Benin). The procedure is the same one. Indeed, it is necessary to

purpose of the microbiological analyses are to highlight weigh 25 grams of the sample, to supplement to 250

the presence or not of fecal coliforms, specifically grams with Peptoned Plug Water (EPT), to incubate

Escherichia coli (E.coli) and salmonellas in the with 37°C during 18 hours, to sow 0,1 ml of this pre-

47 Dougnon et al.
 Int. J. Biosci. 2012

enrichment in 10 ml of medium Rappaport Vasiliadis

and 2 ml of the same pre-enrichment in 20 ml of Water of watering
medium Kaufman, to incubate with 37°C during 24 Water of watering of the witness site of Glo is not
hours ± 4, to insulate the colonies resulting from this contaminated by fecal coliforms (0 UFC/100ml)
enrichment on mediums Xylose Lysine Decarboxylase whereas those of Houeyiho, Fidjrosse and
(XLD) and Hektoen, to incubate with 37°C during 24 Agongbomey are it respectively with 14.5.104 ± 70.06
hours ± 4, to select the characteristic colonies of UFC/100 ml; 8.5.104± 106.01 UFC/100 ml and
salmonellas, to carry out the test of discrimination by 1.05.105± 70.06 UFC/100 ml. Escherichia coli misses
sowing the urea medium and waiting for two hours. on the witness site of Glo contrary with Houeyiho
The salmonellas don’t have urease. (1.8.104 ± 28.42 UFC/100 ml), with Fidjrosse (0.95.104
± 70.10 UFC/100 ml) and with Agongbomey (5545 ±
Physicochemical analyses 77.53 UFC/100 ml) (Fig. 2). These differences prove to
The physicochemical analyses were carried out in the be significant since it is about total absence on the site
Analysis laboratory of the Ground of the Agronomic of Glo and presence of coliforms on the other sites (p<
Faculty of Science, in Republic of the Benin. It was 0.05).
about the determination of the pH. The measurement
of the pH of the ground and the droppings informs
about their degree of acidity or of alkalinity. With this
intention, and in accordance with the standard of
reference NF ISO 10390: 1994, the samples were
filtered with a sieve of mesh 0,2 mm. Twenty grams of
each sample were weighed in bechers and we add it 50
ml of distilled water which and agitated during 15
minutes using an agitator. The solution obtained was
left at rest during thirty minutes. The reading of the pH Fig. 2. Presence of fecal coliforms in water of
was made using a pH-meter of straw mattress CG 825. watering.
For the water samples, the determination was done
directly using the pH-meter of straw mattress CG 825
of manner in conformity with the standard of reference
NF T 90-008: 2001.

Statistical analyses
It was calculated the averages and the standard
deviations. Multiple comparisons consisting in
comparing the averages using the test of Student p
(T>t) = 0.05 were made. The softwares used are Fig.3. Variation of the grounds pH according to site.
Microsoft Excel 2010 and XL Stat 2011.

Results Vegetable leaves

Presence of fecal coliforms in the culture’s The leaves of Solanum macrocarpum L. of the witness
environment of Solanum macrocarpum L. site of Glo are contaminated by fecal coliforms

48 Dougnon et al.
 Int. J. Biosci. 2012

(6.05.105 ± 219.10 UFC/g) whereas those of Houeyiho, Contamination by the salmonellas of the environment
Fidjrosse and Agongbomey are it respectively at a rate of culture of Solanum macrocarpum L.
of 0.95.103 ± 70.71 UFC/g; 3.5.103 ± 21.32 UFC/g and The salmonellas were not detected in the environment
182.103 ± 25.01 UFC/g (Table 1). of culture of Solanum macrocarpum L. Indeed, it was
observed 0 UFC/100 ml in all the samples of grounds,
Table 1. Presence of fecal coliforms in the leaves of of water of watering and the leaves of Solanum
Solanum macrocarpum L. according to sites. macrocarpum L. resulting from the sites of study.
Sites G H F A
Samples of 4.5.105 1.103 5.103 4.103 Variation of the grounds pH
S.macrocarpum
leaves 7.6.105 0.9.103 2.103 3.6.105 With regard to the pH, the averages obtained are as
Average 6.05.105 0.95.103 3,5.103 182.103 follows: Glo (7.305 ± 0.064); Houeyiho (6.550 ±
±Standard 219.10a 70.71a 21.32a 25.01a 0.014); Agongbomey (6.445± 0.120) (p< 0.05);
deviation
Fidjrosse (6.480 ± 0.311) (Fig. 3).
G, H, F and A are respectively the initials of Glo, Houeyiho,
Fidjrosse and Agongbomey. The averages carrying the same
letters are not significantly different with the threshold from Discussion
significance α = 0.05. Presence of fecal coliforms in the environment of
culture of Solanum macrocarpum L.
Table 2. Presence of Escherichia coli in the leaves of It was required the fecal coliforms whose Escherichia
Solanum macrocarpum L. according to sites. coli in water of watering and the leaves of Solanum
Escherichia coli (UFC/100 ml) macrocarpum L. because these bacteria are produced
Sites G H F A
in the intestine of the animals. The ground also
Samples of 3.3.104 1.103 1.103 0.9.103
leaves constitutes according to Pilet (1981) their natural
6.8.10 5 0.9.10 3 1.10 3 0.9.10 3 habitat. But their presence in vegetables is abnormal
Average 35.65.104a 0.95.103a 1.103b and accounts for the sanitary quality of this food. The
0.9.103b
± Standard 457,08 70,71 0 0 presence of fecal coliforms brought by our results can
deviation
be an indication of the presence of micro-organisms
G, H, F and A are respectively the initials of Glo, Houeyiho,
like one notified Zmirou et al. (1987). Moreover, the
Fidjrosse and Agongbomey. The averages carrying the same
presence of Escherichia coli confirms the effective
letters are not significantly different with the threshold from
presence of the fecal coliforms. Also, Habtesealami et
significance α = 0.05.
al. (2010) stress that Escherichia coli develops close to
the roots of the plants and can contaminate the culture
The contamination of the leaves of Solanum
of the young growths. It can live during weeks around
macrocarpum L. by Escherichia coli is of 35.65.104 ±
the roots of the plants and be transferred towards the
457.08 UFC/g on the witness site of Glo whereas it is
edible parts.
respectively of 0.95.103 ± 70.71 UFC/g; 1.103 ± 0
UFC/g and 0.9.103± 0 UFC/g on the sites of Houeyiho,
Generally, the water of watering used on the market-
Fidjrosse and Agongbomey. The statistical analyses
gardening sites is contaminated by the fecal coliforms
revealed that the difference between the averages of
except for the witness site of Glo. This difference could
Glo and Houeyiho is not significant with the threshold
be explained by the system of watering used. Indeed,
of significance α = 0.05. On the other hand, it is it with
on the site of Glo, the market-gardeners use the tap of
the same threshold between the averages of Glo and
the National Company of Water of Benin whereas on
Fidjrosse then Glo and Agongbomey (Table 2).
the other sites, the ground systems of marsh dug

49 Dougnon et al.
 Int. J. Biosci. 2012

(Houeyiho and Agongbomey) and of drilling affirmed that the fecal rate of coliforms in vegetables
(Fidjrosse) are adopted. Water of watering of Houeyiho decreases considerably if the composting of the
is polluted in fecal coliforms than those of droppings used to amend them is adequate. In
Agongbomey and Fidjrosse. The results of this study addition, the leaves coming from all the sites present
are in agreement with those of Akodogbo (2005) which values higher than those imposed by the provincial
stress that the water of drilling is polluted than that of Committee of Canada on the standardization and the
the wells (grounds dug of marsh). This author interpretation of the microbiological criteria in the
remarked in addition that the majority of the well food which is of 1,0.102 UFC/g (Santé Canada, 1991).
rivers and drilling of Cotonou are invaded by the
coliforms. Presence of the salmonellas in the culture’s
environment of Solanum macrocarpum L.
It is the same for Escherichia coli absent on the site of Generally, no salmonella was found in the
Glo but present on the other market-gardening sites. environment of culture of Solanum macrocarpum. The
That confirms well this water of watering knew a absence of these germs in the water of watering, the
recent fecal contamination. This is possible insofar as ground and the leaves of Solanum macrocarpum L.
the majority of the aforesaid sites have a considerable could mean there’s no contamination of the droppings
insalubrity (human excrements and other waste). The used for the amendment of this culture. Similar results
depth of the wells is also a factor very significant were reported by Araba et al. (2000) which sought the
because more water infiltrates, more it gets rid of its salmonellas in nine samples of chicken droppings.
impurities; however the majority of the sources of Indeed, the enterobacteria are naturally present in the
watering of the market-gardening sites are located in small intestine of the animals; their presence in a
hollows and are not deep. None water of watering, put medium always has a fecal origin (Pilet, 1981).
besides those of the site of Glo meets the standards
varying between 100 and 200 UFC/100 ml for the However, Rose et al. (1999) identified factors of risk
coliforms and Escherichia coli, proposed by Santé relating to the contamination at 70% by Salmonella in
Canada (1991). the chicken droppings. This difference could be related
to the taking away of droppings coldly collected by
The leaves of Solanum macrocarpum L., indeed, are these authors on the level of the hen houses contrary to
contaminated by the fecal coliforms on all the sites of the method of collection adopted during the present
study. This contamination would be certainly due study. Indeed, the chicken droppings were obtained
either to the microbiological quality of the water of after transport and storage during several weeks before
watering used on these sites, or with the chicken their use. Storage aims in particular to cleanse the
dropping being used for the amendment. The presence manure, by the heating which produced there. Kwak et
of coliforms on the witness site of Glo, amended by the al. (2005) have for this purpose, evaluated the effects
chicken dropping and where water of watering is those of an in core treatment deep of the chicken litter. They
of tap could accuse the droppings of chickens like thus suggested that the correctly piled up poultry litter,
principal source of contamination. Indeed, the with or without ventilation, ensures the elimination of
vegetable leaves resulting from the pilot site showed a the pathogenic enterobacteria in 8 days. Moreover,
strong rate of contamination by the fecal coliforms Araba et al. (2000) bound the absence of salmonellas
because the droppings used did not know any in their study with storage carried out at the level of
composting before the amendment of the seedlings. It the farms preceding the microbiological examinations.
confirms the results of Florin et al. (2009) which

50 Dougnon et al.
 Int. J. Biosci. 2012

However, it could be that the droppings used for the Akodogbo H. 2005. Contribution à l'amélioration
present study results from farms which were free of de la qualité de l'eau à usage domestique dans le 5ème
salmonella’s contamination. arrondissement de la commune de Porto-Novo au
Bénin, Mémoire de Maîtrise Professionnelle en
The pH and Escherichia coli Environnement et Santé, FLASH/UAC, p. 65.
Within the framework of this study, the ground of the
witness site of Glo has a pH of 7,305, very near to Araba A, Hachana Y. 2000. Caractérisation
neutrality. It is noticed that Escherichia coli strongly chimique et microbiologique des fientes de volaille en
developed on the leaves resulting from this site vue de leur utilisation en alimentation des ruminants,
contrary to the other sites whose pH is basic. Actes Institut Agronomique et Vétérinaire du Maroc,
Enterobacteria such that Escherichia coli develop with 20(2), 83-88
an optimum pH near to the neutrality which varies
between 7, 2 and 7, 4 like remarked by Delahaye Chidikofan G. 2010. Contribution à l’amélioration
(2009). In addition, the alkalinity of the grounds of de la qualité des cultures maraîchères du site de
Houeyiho and Agongbomey compared to that of the Houéyiho à Cotonou au BENIN : cas de la laitue
other sites could be explained by the seniority of the (Lactuca sativa L.), Mémoire de Master, 2IE
site; exploitation of the sites of Glo and Fidjrosse being Ouagadougou, p. 56.
more recent.
Delahaye A. 2009. La famille des entérobactéries,
Conclusion disponible sur http://www.arnobio.com, Consulté le
The results of this study indicate that no risk of 18/07/2011
contamination of the leaves of Solanum macrocarpum
L. by the salmonellas was detected on the sites of Florin NH., Maddocks AR., Wood S., Harris
production. On the other hand, some risks of AT. 2009. High-temperature thermal destruction of
contamination of water of watering by the fecal poultry derived wastes for energy recovery in Australia.
coliforms whose Escherichia coli were identified on the Waste Management 29(4), 1399-1408
market-gardening sites except that of Glo on the one
hand and some risks of contamination of the leaves of Habteselassie-Mussie Y, Bischoff M, Applegate
Solanum macrocarpum L. by the fecal coliforms and B, Reuhs B, Turco Ronald F. 2010.
Escherichia coli were detected on all the studied sites Understanding the role of Agricultural Practices in the
on the other hand. The presence of pathogenic micro- Potential Colonization and Contamination by
organisms on vegetables of the various sites can be at Escherichia coli in the rhizospheres of Fresh Produce.
the origin of food poisonings; this forecasts new Journal of Food Protection 73, 2001-2009.
prospects for research in order to cleanse vegetables
consumed by the populations of Cotonou. Koc M, Macrae R, Mougeot JAL, Welsh J.
2000. Armer les villes contre la faim: systèmes
References alimentaires urbains durables. CRDI p. 243.
Akinbamijo OO., Fall ST., Smith OB. 2002.
Advances in crop-livestock integration in West African Kwak WS, Huh JW, Maccaskey TA. 2005. Effect
cities. Dakar: ITC/ISRA/CRDI, p. 213. of processing time on enteric bacteria survival and on
temperature and chemical composition of broiler

51 Dougnon et al.
 Int. J. Biosci. 2012

poultry litter processed by two methods. Bioresource Rose N, Beaudreau F, Drouin P, Toux JY, Colin
Technology 96(14), 1529-1536. RP. 1999. Risk factors for Salmonella enterica subsp.
enterica contamination in French broiler-chicken
Legret M, Divet L, Juste C. 1988. Migration et flocks at the end of the rearing period. Preventive
spéciation des métaux lourds dans un sol soumis à des Veterinary Medicine 39(4), 265-277.
épandages boues de station d’épuration à très forte
charge en Cd et Ni. Wat. Res. 22(8), 953-959. Santé C. 1991. La qualité bactériologique. Document
de support aux « recommandations pour la qualité de
Métras R. 2003. Utilisations et dangers sanitaires l’eau potable au Canada ». Accessible à : www.hc-
microbiologiques liés aux effluents d’élevage, Thèse de sc.gc.ca/ehp/dhm/dpc_eau_qualite/eauguide.htm
doctorat unique, Université Claude Bernard p. 150.
Smit J, Ratta A, Bernstein J. 1996. Urban
Mougeot J. 2006. Focus: cultiver de meilleures agriculture: an opportunity for environmentally
villes. Agriculture urbaine et développement durable, sustainable development in sub-Saharan Africa,
CRDI, 136p. Washington (DC, E.-U.), Banque mondiale,
Environmentally Sustainable Division, African
MVAD. 2003. Les matières organiques produites à la Technical Department. Post-UNCED Series, Building
Réunion, chapitre 5, p. 63. Blocks for Africa 2025, Paper No. 11.

Niang S. 1996. Utilisation des eaux usées Zmirou D, Ferley JP, Collin JF, Charrel M,
domestiques en maraîchage périurbain à Dakar Berlin J. 1987. A follow-up study of gastro-intestinal
(Sénégal). Sécheresse 3, 217-23 diseases related to bacteriologically substandard
drinking water. American Journal of Public Health 77,
Pilet C, Bourdon L, Toma B, Marchal N, 582-584.
Balbastre C. 1981. Bactériologie médicale et
vétérinaire, systématique bactérienne, Doin éditeurs-
Paris ; 2ème édition, 2ème tirage, p. 431.

52 Dougnon et al.