Thesis
Occurrence and Dissemination of Micropollutants and Antibiotic Resistance in Aquatic Environment: A Prevalence Study across Geographical Location and Different Systems of Wastewater Management
LAFFITE, Amandine
Abstract
The pollution of water is a major problem in many parts of the world. In many developing countries, aquatic systems are receiving untreated or partially treated effluents, containing anthropogenic pollutants whereas these rivers serve as a basic network for human and animal consumption. High values of toxic metals, persistent organic pollutants (POPs), faecal indicator bacteria (FIBs), antibiotic resistant bacteria (ARBs) and antibiotic resistance genes (ARGs) in rivers may pose a great risk to human health and aquatic living organisms. The main objective of the research is to assess the prevalence and dissemination of toxic metals, POPs, FIB, ARB and ARGs in the rivers of Kinshasa (Republic Democratic of the Congo) as a study case. Overall, the present work demonstrated that chemical and microbiological pollution can exceed, in many studied sites, the international recommendation for water quality and has the potential to affect ecosystem functions as well as human impact.
Reference
LAFFITE, Amandine. Occurrence and Dissemination of Micropollutants and Antibiotic Resistance in Aquatic Environment: A Prevalence Study across Geographical Location and Different Systems of Wastewater Management. Thèse de doctorat : Univ. Genève, 2019, no. Sc. 5355
DOI : 10.13097/archive-ouverte/unige:120648 URN : urn:nbn:ch:unige-1206485
Available at: http://archive-ouverte.unige.ch/unige:120648
Disclaimer: layout of this document may differ from the published version.
1 / 1 UNIVERSITÉ DE GENÈVE FACULTÉ DES SCIENCES
Section des Sciences de la Terre et de l’Environnement Département F.-A. Forel des sciences de l’environnement et de l’eau Dr. John Poté Prof. Vera Slaveykova
Occurrence and Dissemination of Micropollutants and Antibiotic Resistance in Aquatic Environment: A Prevalence Study across Geographical Location and Different Systems of Wastewater Management
THÈSE
présentée à la Faculté des Sciences de l’Université de Genève pour obtenir le grade de Docteur ès Sciences, mention Science de l’Environnement
par
Amandine LAFFITE
de
Perpignan (France)
Thèse N°5355
GENÈVE 2019
UNIVERsITÉ DE CENÊVE
FÂ{{JN-TÉ f}fis SflIEruf KS
DOCTORAT ES SCIENCES, MENTION SCIENCES DE L'ENVIRONNEMENT
Thèse de Madame Amandine LAFFITE
intitulée :
<
La Faculté des sciences, sur le préavis de Monsieur J. POTE-WEMBONYAMA, docteur et directeur de thèse (Département F.-A. Forel des sciences de I'environnement et de I'eau), Madame V. SLAVEYKOVA, professeure ordinaire et codirectrice de thèse (Département F.-A. Forel des sciences de l'environnement et de I'eau), Monsieur B. W. IBELINGS, professeur ordinaire (Département F.-A. Forel des sciences de I'environnement et de l'eau), Monsieur S. J. HARBARTH, professeur ordinaire (Faculté de médecine, Service de prévention de I'infection) et Monsieur C. MULAJI KYELA, professeur (Département de chimie, Université de Kinshasa, République démocratique du Congo), autorise I'impression de la présente thèse, sans exprimer d'opinion sur les propositions qui y sont énoncées.
Genève, le 6 juin 2019
Thèse - 5355 -
Le Doyen
N.B. - La thèse doit porter la déclaration précédente et remplir les conditions énumérées dans les "lnformations relatives aux thèses de doctorat à I'Université de Genève".
Table des matières
Abstract ...... VII
Résumé ...... XIII
Abbreviation lists ...... XVII
Figure caption ...... XXI
Table caption ...... XXV
...... 1
1.1 Anthropogenic impact on aquatic systems ...... 3
1.1.1 Contamination sources ...... 3
1.1.2 Different types of pollutants ...... 4
1.1.3 Contamination variability ...... 7
1.1.4 Wastewater management systems ...... 8
1.2 Antibiotic resistance genes as emerging pollutant ...... 10
1.2.1 A brief overview of antibiotic resistance ...... 11
1.2.2 General mechanisms of antibiotic resistance ...... 12
1.2.3 Β-lactams and β-lactam resistance ...... 14
1.2.4 Genetic support and transfer mechanisms of antimicrobial resistance ..... 17
1.2.5 Link between AMR and virulence ...... 21
1.2.6 The global spread of antibiotic resistance and the rise of superbugs ...... 22
1.2.7 Relevance of antibiotic resistance in the environment ...... 23
1.3 Research objectives and chapters’ organization ...... 26
1.3.1 Research objectives ...... 26
1.3.2 Thesis outline ...... 27
1.3.3 Study site implicated in the research ...... 27
1.3.4 Institutional framework ...... 30
1.3.5 Funding ...... 31
I
1.3.6 List of publication relative to this research ...... 31
References ...... 34
...... 41
Abstract ...... 42
2.1 Introduction ...... 43
2.2 Material and methods ...... 44
2.2.1 Study sites and sampling procedure ...... 44
2.2.2 Sediment grain size and organic matter, total organic carbon, total nitrogen and phosphorus analysis ...... 45
2.2.3 Metal analysis in sediment samples ...... 46
2.2.4 Geoaccumulation index and enrichment factor ...... 46
2.2.5 Chlorinated pesticides, PCBs, PAHs and PBDEs analysis ...... 47
2.2.6 Data analysis ...... 49
2.3 Results and discussion ...... 49
2.3.1 Physicochemical characteristics of sediments ...... 49
2.3.2 Metal concentrations in the surface sediments ...... 51
2.3.3 Enrichment factor (EF) and Geoaccumulation index (Igeo) ...... 52
2.3.4 Spatial distribution of persistent organic pollutants in sediments ...... 54
2.3.5 Correlation between parameters ...... 63
2.4 Conclusion ...... 64
References ...... 65
...... 71
Abstract ...... 72
3.1 Introduction ...... 73
3.2 Materials and Methods ...... 75
II
3.2.1 Study site description ...... 75
3.2.2 Sampling procedure ...... 76
3.2.3 Water physicochemical parameter analysis ...... 78
3.2.4 Faecal indicator bacteria (FIB) analysis in water and sediment samples . 78
3.2.5 Characterization of FIB strains ...... 78
3.2.6 Data Analysis ...... 79
3.3 Results and discussion ...... 80
3.3.1 Water physicochemical parameters ...... 80
3.3.2 River microbiological quality ...... 81
3.3.3 Microbiological quality of water from wells ...... 84
3.3.4 Characterization of Faecal Indicator Bacteria ...... 85
3.3.5 Correlation between parameters ...... 86
3.4 Conclusion ...... 87
References ...... 89
..... 89
Abstract ...... 96
4.1 Introduction ...... 97
4.2 Material and Method ...... 100
4.2.1 Study site and sampling ...... 100
4.2.2 Sediment grain size, organic matter and water content ...... 101
4.2.3 Toxic metal analysis ...... 101
4.2.4 Total DNA extraction ...... 102
4.2.5 qPCR Quantification of selected genes in sediments: 16S rRNA, ARGs and FIB ...... 102
4.2.6 Data analysis ...... 103
4.3 Results and Discussion ...... 104
4.3.1 Sediment physicochemical parameters and metal content ...... 104
III
4.3.2 Abundance of bacterial population ...... 107
4.3.3 Quantification of antibiotic resistance genes ...... 111
4.3.4 Statistical correlation ...... 114
4.4 Conclusion ...... 115
References ...... 117
125
Abstract ...... 126
5.1 Introduction ...... 127
5.2 Material and Methods ...... 128
5.2.1 Study sites and sampling procedures ...... 128
5.2.2 Sediment physico-chemical parameters: sediment grain size, total organic matter and carbonates ...... 128
5.2.3 Toxic metal analysis in sediment samples ...... 129
5.2.4 Geoaccumulation index (Igeo), erichment factor (EF), and single pollution index (PI) ...... 129
5.2.5 ARGs quantitation by qPCR ...... 130
5.2.6 Statistical analysis ...... 131
5.3 Results ...... 132
5.3.1 Sediment Physicochemical Parameters ...... 132
5.3.2 Sediment metal content and toxic metal pollution assessment ...... 133
5.3.3 Abundance of antibiotic resistance genes in urban rivers ...... 136
5.3.4 Correlation between chemical and biological parameters ...... 137
5.4 Discussion ...... 140
5.5 Conclusion ...... 142
References ...... 143
IV
149
Abstract ...... 150
6.1 Introduction ...... 151
6.2 Material and Methods ...... 153
6.2.1 Study site and sampling settings ...... 153
6.2.2 Escherichia coli isolation and antimicrobial susceptibility testing ...... 154
6.2.3 Determination of strains phylogenetic groups ...... 154
6.2.4 Virulence factors and pathogenic islands ...... 156
6.2.5 ESBL identification and antimicrobial susceptibility testing ...... 157
6.2.6 Statistical analysis ...... 157
6.3 Results and Discussion ...... 157
6.3.1 Genotyping and phylogenetic grouping of isolates ...... 157
6.3.2 Phenotypic resistance ...... 158
6.3.3 Distribution of antimicrobial resistance genes and mobile genetic elements in ESBLEC isolates ...... 160
6.3.4 Association of virulence genes, DEC pathogenic islands and antibiotic resistance pattern among phylogroups ...... 162
6.4 Conclusion ...... 163
References ...... 165
...... 171
Abstract ...... 172
7.1 Introduction ...... 173
7.2 Materials and Methods ...... 174
7.2.1 Site description and sampling ...... 174
7.2.2 Sediment physicochemical characterization ...... 177
7.2.3 Toxic metal analysis ...... 177
V
7.2.4 ARGs quantification by qPCR ...... 178
7.2.5 Data analysis ...... 178
7.3 Results and discussion ...... 179
7.3.1 Sediment physicochemical characteristics and metal content ...... 179
7.3.2 Quantification of bacterial population (16S rRNA) ...... 181
7.3.3 Quantification and contamination of antibiotic resistant genes ...... 182
7.3.4 ARGs and physicochemical parameter correlations ...... 185
7.4 Conclusion ...... 187
References ...... 188
...... 195
8.1 Conclusions ...... 196
8.1.1 Dissemination of metals and POPs in urban rivers of Kinshasa ...... 196
8.1.2 Surface water contamination by FIB linked to poor management of wastewaters ...... 196
8.1.3 Organic and inorganic pollution in urban river receiving untreated effluents ...... 197
8.1.4 Prevalence of ESBLs and Carpapenem resistance genes in rivers receiving untreated effluents ...... 197
8.1.5 Virulence and AMR properties of ESBLs ...... 198
8.1.6 Influence of the wastewater management on the dissemination of micro- and emerging pollutants ...... 198
8.2 Perspectives ...... 199
VI
Remerciements
Le cours de la vie est surprenant et captivant. Au détour des nombreuses aventures, on à la joie de rencontrer de fantastiques personnes. Aujourd’hui, je remercie toutes ces personnes qui par leur bienveillance (voire même par leur malveillance) m’ont donné le courage de partir loin de « mon chez moi » seule et avec un enfant sous le bras, ainsi que la force et la ténacité nécessaire à l’accomplissement de cette thèse avec la volonté de prouver que tout est possible.
Je voudrais remercier le Fond National Suisse pour le financement de ces travaux de recherche.
Je remercie les membres du jury, les professeurs Stephan Harbarth, Bastiann Ibelings et Crispin Mulaji pour avoir accepté d’évaluer mes travaux.
Je souhaite exprimer ma sincère gratitude à mes superviseurs, le Dr John Poté et le Prof. Vera Slaveykova pour m’avoir guidé le long de ces 5 années et m’avoir donné l’opportunité de découvrir ces deux magnifiques pays que sont la Suisse et le Congo RD. La Mundélé que je suis, garde de fabuleux souvenir de la RDC (j’essaie quand même d’effacer les parties énormissimes moustiques et lavage de cheveux avec les seaux d’eaux à température locale). Je les remercie d’avoir toujours été là tout en me laissant l’autonomie de développer mes idées de recherche ainsi que de m’avoir laissé apprendre à gérer les besoins de laboratoire et la supervision des stagiaires. Ces 5 années auprès de vous ont été tellement formatrices.
Je remercie tous les Foréliens de Versoix/ Carl Vogt que j’ai eu la chance de rencontrer au cours de ces dernières années. Grâce à vous, je n’ai jamais eu autant de caféine dans le sang au cours de ma vie ! Severine Le Faucheur, Carmen Moinecourt, Claudia Cosio, le Bruxelles 2017 avec vous restera inoubliable ! Dhafer Al Salah, Giulia Cheloni, Kilian Kavanagh, Wei Liu, Isa Worms, Marianne Seijo, Sebastien Guilmot, Teofana Chonova, Fred Loosli, que de bon moments passés avec vous tous ! Alexandra Baeriswyl Beuchat, Katia Loizeau, Philippe Arpagaus, je vous remercie pour le support technique, administratif et informatique apporté. Vous avez été d’un grand secours.
Je remercie ma famille un peu partout en France pour son soutien indéfectible et pour les longues heures passées au téléphone quand le moral n’était plus tellement là. Et plus particulièrement à mes grand parents Yoyo et JP pour s’être toujours dévoués pour venir garder les enfants pendant les moments difficiles et m’avoir hébergée pendant une partie de mes années lyonnaises. A mon oncle Rodolphe qui a vu sa tranquillité de célibataire endurci dévastée par
VII l’arrivée d’une étudiante et de son bébé Flavien de 9 mois dans son appartement pendant mes années de master. A mon Mac Gyver personnel papa Michel Laffite pour m’avoir aidé à soigner à distance mes bobos de doctorat (ordinateur dans le yaourth, autoclave qui ne ferme plus, moral à la baisse et j’en passe…). Je vous aime tellement !
Pour terminer, qui aurait cru que faire prendre mes affaires de moto par un blablacar, m’aurait permis de te re-rencontrer et que de parler de la rue Princesse ne t’aurais pas fait fuir… Jonathan Moreau tu es la plus belle chose qui me soit arrivée dans la vie. Je te remercie de supporter ma folie de scientifique au quotidien. Je pense que ça ne doit pas être facile tous les jours de supporter mes excentricités. Je t’aime mon chéri ! Et les plus gros des câlins vont à mes petits amours Flavien et Roxane, les petits piments de vie qui me mettent au défi tous les jours et me rappellent tous les jours que les petites choses de la vie sont aussi les plus beaux moments à vivre.
VIII Abstract
The pollution of water resources by different types of contaminants, including inorganic and organic micropollutants (toxic metals, polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs)), faecal indicator bacteria (FIB), pathogenic organisms, antibiotic resistant bacteria (ARB) and their resistance genes (ARGs) is a major problem in many parts of the world. However, the nature of contaminants and the degree of pollution depends on both degree of country development and the system of wastewater management. In developed countries, industrial and municipal wastewater treatment plants (WWTP) and agricultural runoff constitute the major sources of contaminants in the aquatic environment. In many developing countries, such as in Sub-Saharan Africa, south of Asia and Central and Latin America, most rivers, lakes, and lagoons are receiving untreated or partially treated urban and industrial effluents, storm water runoff containing anthropogenic pollutants due to intensive and uncontrolled urbanization. The water resource pollution may be a result of multiple diffuse pollution sources including open defecation, uncontrolled landfills, untreated hospital effluents, unregulated urban effluents discharge, and inadequate sewage collection. Rivers in most developing nations serve as a basic network for human and animal consumption as well as irrigation for fresh urban produces. High values of toxic metals, persistent organic pollutants (POPs), FIBs, ARBs and ARGs in rivers may pose a great risk to human health and aquatic living organisms.
The main objective of the present interdisciplinary research is to assess and to perform a comparative survey on the prevalence and dissemination of toxic metals, POPs, FIB, ARB and ARGs: broad-spectrum β-lactam and sulfonamide resistance genes (blaSHV, blaCTX-M, blaNDM, blaKPC, blaOXA-48, blaVIM, blaIMP, sul1 and sul2) and the total bacterial load (16S rRNA genes). Several receiving systems with respect to degree of wastewater management were selected including different receiving systems in Kinshasa (such as the rivers Makelele, Kalamu, Kokolo Canal and Nsanga), the capital city of Democratic Republic of the Congo. The rivers in Kinshasa receive urban, hospital and industrial effluents without any previous treatment and serve as sources of water for many uses such as recreational activities, bathing, drinking water supply and irrigation for urban agriculture. A comparison is also made with Lake Brêt in Switzerland, that is not affected by WWTP effluents.
In the river receiving systems of Kinshasa, high concentration of toxic metals (Cr, Cu, Zn, Cd, Pb, and Hg) was observed in many sampling sites reaching the values (in mg kg-1) of
IX
47.9 (Cr), 213.6 (Cu), 1434.4 (Zn), 2.6 (Cd), 281.5 (Pb), and 13.6 (Hg). These concentrations largely exceed the sediment quality guidelines (SQGs) and the probable effect level (PEL) for the protection of aquatic life recommendation. High values of PCBs and OCPs were detected in sediment samples, e.g. Makele and Nsanga rivers, PCB values ranged from 0.9-10.9 with total PCBs (Σ7PCBs x4.3): 169.3 µg kg-1; OCPs from 21.6-146.8 with ΣOCPs: 270.6 µg kg-1. The polybrominated biphenyl ethers (PBDEs) concentrations were higher in investigated rivers comparatively with values detected in many rivers from Sub-Saharan Africa. The PAHs value ranged from 22.6 to 1011.9 µg kg-1. River contamination may be explained by local intense domestic activities, urban and agricultural runoff, industrial and hospital wastewaters discharge into the rivers without prior treatment. These river receiving systems are also characterized by contamination of pathogens, virulence genes, FIBs, ARBs and ARGs varying according to the season. The pollution level was significantly higher during the wet season compared with the dry season. For example in Kokolo Canal, during wet and dry season, E. coli reached the values of 18.6x105 and 4.9x105 CFU 100 mL-1 and Enterococcus reached the values of 7.4x104 and 2.7x104 CFU 100 mL-1. Strong mutually positive correlation was observed between E. coli and ENT, with the range of R-value being 0.93 < R < 0.97 (p-value < 0.001, n=15). The PCR assays for human-specific Bacteroides indicated that more than 98% of 500 isolated FIB strains were of human origin, pointing out the effect of poor household sanitation practices on surface water but also on groundwater contamination. The influence of the hospital effluents discharge into the receiving systems was illustrated only for two of the four studied hospital by for the presence and dissemination of -1 ARBs and ARGs. The ARGs copy number (log10 ARGs g of dry sediment) for aadA, blaCTX-
M, blaSHV, and blaTEM varied respectively from 4.64 to 7.83, 4.67 to 5.01, 3.92 to 4.66, and 4.23 to 4.86 at upstream sites, and from 5.33 to 9.24, 4.55 to 5.61, 3.76 to 6.17, and 4.36 to 5.46 at downstream sampling sites of three studied hospital outlet effluents discharge. The abundance of ARGs and associated bacteria in a receiving system varied depending on the size of the hospital and type of services. The abundance of total bacterial load, ARGs, and FIB showed a relevant increase after hospital wastewater discharge but the results highlight that hospital wastewater effluents are not the only source of these pollutants in the sediment of studied river receiving systems.
In Lake Brêt, the results demonstrated a widespread dissemination of blaTEM, blaSHV and sul1, which were also highly correlated to bacterial biomass and organic matter content. These findings demonstrate a fixation of last ARGs generation in the environment whereas actual antibiotic regulation tend to limit the dissemination of other ARGs in the studied lake reservoir. X
On the other hand, comparing to the Bay of Vidy, which receive WWTP effluent waters, the Lake Brêt seems to be less affected by broad-spectrum β-lactams and other resistance genes linked to antibiotics currently used in human and veterinary medicine. This study reveal that the anthropogenic pressure is a major driver of pollutant dissemination and that WWTP outlet is linked to pollutant dissemination in aquatic system.
Overall, the present work demonstrated that chemical and microbiological pollution can exceed, in many studied sites, the international recommendation for water quality and has the potential to affect the ecosystem functions as well as human impact. The quantification of toxic metals, POPs, FIB, ARB and ARGs as performed in this study contributes for the improvement of the risk assessments as well as provides baseline information for developing strategies (such as hospital and urban wastewater treatment, agricultural runoff) to limit the spread of these contaminants in different aquatic environmental settings.
XI
Résumé
La pollution des ressources en eau par des polluants variés, dont les micropolluants organiques et inorganiques (métaux toxiques, hydrocarbones aromatiques polycycliques (HAPs), biphényls polychlorés (PCBs) et pesticides organochlorés (OCPs)), les bactéries indicatrices de contamination fécale (FIBs), les microorganismes pathogènes, les bactéries résistantes aux antibiotiques (ARBs) et leurs gènes de résistance (ARGs) associés, sont encore un problème majeur dans le monde. Cette problématique dépend à la fois du degré de développement du pays et de leur système de gestion des eaux usées. Dans les pays développés, les stations d’épuration des eaux usées municipales et industrielles (STEP) et le ruissellement des sols agricoles sont les responsables majeurs de la contamination des systèmes aquatique. La situation est encore plus alarmante dans les pays en voie de développement tel que l’Afrique Sub-Saharienne, l’Asie du sud ainsi que les Amériques Centrales et Latines. Dans ces pays, les systèmes aquatiques adjacents tel que les lacs, les rivières et les lagons reçoivent des eaux polluées par l’urbanisation massive et incontrôlée. La pollution des ressources en eau peut être expliquée par l’additivité des pollutions multidiffuses incluant la défécation à l’air libre, les décharges incontrôlées, les effluents hospitaliers non traités, le rejet d’effluents urbains non règlementés ainsi que la collecte inadéquate des eaux usées. Dans la plupart des pays en voie de développement, les rivières sont utilisées pour la consommation en eau humaine et animale ainsi que pour l’irrigation des produits maraîchers. Une forte contamination en métaux toxiques, polluants organiques persistants (POPs), FIB, ARB et ARGs a été détecté dans les rivières ; contamination qui pourrait poser un risque majeur pour la santé humaine et celle des organismes aquatiques.
L’objectif principal de cette recherche pluridisciplinaire est d’effectuer et d’évaluer une étude comparative sur la prévalence et la dissémination des métaux toxiques, POPs, FIBs, ARBs et ARGs : gènes de β-lactamases à spectre étendu et de résistance aux sulfonamides
(blaSHV, blaCTX-M, blaNDM, blaKPC, blaOXA-48, blaVIM, blaIMP, sul1 and sul2) en lien avec la biomasse bactérienne totale (16S rRNA). Des systèmes aquatiques receveurs ont été sélectionnés en fonction du niveau de traitement des eaux usées. Ces sites incluent des systèmes receveurs à Kinshasa (tel que les rivières Makelele, Kalamu, le Canal de Kokolo et Nsanga), la capitale de la République Démocratique du Congo. Les rivières de Kinshasa reçoivent des effluents urbains hospitaliers et industriels sans aucun traitement préalable et ces rivières servent par la suite de source d’eau pour de nombreuses utilisations tel que les activités
XIII récréatives, la baignade, l’eau potable et l’irrigation du maraîchage urbain. Une comparaison a été effectuée avec le lac de Brêt (Suisse) ; lac qui n’est pas affecté par les effluent de STEP.
Dans les rivières urbaines de Kinshasa, de très fortes concentrations en métaux toxiques (Cr, Cu, Zn, Cd, Pb, and Hg) ont été observé dans la majorité des sites échantillonnés. Ces concentrations peuvent attendre des valeurs (en mg kg-1) jusqu’à 47.9 (Cr), 213.6 (Cu), 1434.4 (Zn), 2.6 (Cd), 281.5 (Pb), et 13.6 (Hg). Ces concentrations dépassent largement les valeurs recommandées (SQGs) et les niveaux d’effet probable (PEL) pour la protection de la vie aquatique. De fortes valeurs de PCBs et d’OCPs ont été détecté dans les sédiments des rivières Makelele et Nsanga. Les valeurs de PCBs sont de l’ordre de 0.9-10.9 avec un total des PCBs (Σ7 PCBs x 4.3) de 69.3 µg kg-1; OCPs:1.6-146.8 avec ΣOCPs: 270.6 µg kg-1. Les concentrations en PBDEs étaient plus élevées dans nos sites d’étude que dans la plupart des rivières d’Afrique Sub-Sahariennes. La PAHs était de l’ordre de 22.6 à 1011.9 µg kg-1. La contamination des rivières par les POPs pourrait être expliquée par l’intensité des activités domestiques et urbaines, par le ruissellement des zones agricoles ainsi que la décharge des eaux usées provenant des industries et des hôpitaux dans les systèmes aquatiques adjacent et ce, en l’absence de prétraitement. Ces systèmes aquatiques sont aussi caractérisés par leur contamination par les pathogènes, les gènes de virulence, les FIB, les ARBs ainsi que leurs ARGs. Le niveau de contamination est variable selon la saison étudiée. Pendant la saison humide, le niveau de pollution mesuré est plus élevé que lors de la saison sèche. Par exemple, dans le canal de Kokolo, les E. coli atteignent les valeurs de 18.6x105 CFU 100 mL-1 et les Enterococcus atteignent des valeurs de 7.4x104 CFU 100 mL-1 pendant la saison humide alors que pendant la saison sèche, ces valeurs étaient de 4.9x105 et 2.7x104 CFU 100 mL-1, respectivement. Une forte corrélation positive a été observée entre les E. coli et les ENT (0.93 < R < 0.97, p-value < 0.001, n=15). Une PCR ciblant la bactérie Bacteroides a déterminé que plus de 98% des 500 isolats FIB étaient d’origine humaine, indiquant que les mauvaises pratiques sanitaires avaient un impact majeur sur la contamination des eaux de surfaces et souterraines.
Dans cette étude, la décharge d’effluents hospitaliers dans le système receveur a démontré un impact sur la dissémination des ARBs et ARGs pour deux des quatres hôpitaux -1 étudiés. Le nombre de copies des ARGs (log10 ARGs g de sediment sec) pour aadA, blaCTX-
M, blaSHV, et blaTEM était respectivement de 4.64 à 7.83, 4.67 à 5.01, 3.92 à 4.66, et 4.23 à 4.86 pour les sites en amont, et de 5.33 à 9.24, 4.55 à 5.61, 3.76 à 6.17, et 4.36 à 5.46 pour les sites en aval des effluents des 4 hôpitaux étudiés. L’abondance bactérienne et des ARGs dans le
XIV système receveur dépendant majoritairement de la taille de l’hôpital ainsi que des services dont il est doté. La biomasse totale, ainsi que l’abondance des ARGs et des FIB a montré une augmentation significative après décharge des effluents, ce qui suggère que les effluents hospitaliers ne sont pas les seuls responsables de la dissémination de l’accumulation des polluants dans les systèmes aquatiques adjacents.
Dans le lac de Brêt, les résultats démontrent la dissémination de blaTEM, blaSHV et sul1. Ces gènes sont aussi fortement corrélés à la biomasse bactérienne et à la teneur en matière organique du sédiment. Ces résultats démontrent la fixation au cours du temps des plus anciens ARGs dans l’environnement alors que la régulation de l’utilisation des antibiotiques en vigueur tend à limiter la dissémination des ARGs les plus récents dans le lac. D’un autre côté, la comparaison avec la Baie de Vidy qui reçoit des eaux traitées par la STEP montre que le lac de Brêt est très peu affecté par les β-lactamases à spectre étendu ainsi que les autres gènes de résistance en lien avec les antibiotiques actuellement utilisés en médecine clinique et vétérinaire. Cette étude révèle que la pression anthropogénique est un acteur majeur de la dissémination des polluants et que le rejet d’effluents traités de STEP sont liés à la dissémination de polluants dans les systèmes aquatiques.
Les résultats de notre étude montrent que les pollutions chimiques et microbiologique peuvent largement dépasser les recommandations internationales pour la qualité de l’eau. Cette pollution peut autant affecter le fonctionnement des écosystèmes que la santé humaine. La quantification des métaux toxiques, POPs, FIBs, ARBs et ARGs effectuée lors de ces travaux de recherches peuvent permettre d’améliorer l’évaluation des risques associés aux rejets d’eaux usée dans les systèmes aquatiques adjacents ainsi que ceux associés à l’utilisation des antibiotiques par l’homme, les animaux et dans l’agriculture.
XV
Abbreviation lists
AFSSA Agence Française de sécurité sanitaire des aliments
APHA American Public Health Association(APHA)
ARB Antibiotic resistant bacteria
ARG Antibiotic resistance genes
BGA Between Group Analysis
BOD5 Biological oxygen demand
BRM Bioreactor membrane
CCME Canadian Council of Ministers of the Environment
CDC Centers for Disease Control and Prevention
CFU Colonies forming unit
CRE Carbapenem-resistant Enterobacteriaceae
DDT Dichlorodiphenytrichloroethane
DNA Deoxyribonucleic acid
E. coli Escherichia coli
EC Electrical conductivity
ENT Enterococcus spp.
ESBL Extending Spectra β-lactamase
FAO Food and Agriculture Organization of the United Nations
FDA U.S. Food and Drug Administration
FIB Faecal indicator bacteria
HGT Horizontal gene transfer
HOC Hydrophobic organic compounds
XVII
ICE Integrative and conjugative elements
INSERM French National Institute of Health and Medical Research
IS Insertion sequence
JMP Joint Monitoring Program 7
MALDI-TOF Matrix Assisted Laser Desorption Ionisation - Time of Flight
MDR Multidrug resistant
MDR-GNB Multidrug Resistant Gram Negative Bacteria
MGE Mobile genetic elements
NAG N-acetylglucosamic acid
NAM N-acetylmuramic acid
NDM New Dehli β-metallolactamase type 1
OCP Organochlorine pesticides
OFAG Federal Office for Agriculture
OFEV Federal Office for the Environment
OSAV Federal Food Safety and Veterinary Office
PAH Polycyclic aromatic hydrocarbons
PAI Pathogenicity islands
PBDE Polybrominated biphenyl ethers
PBP Penicillin Binding Proteins
PCB Polychlorinated biphenyls
PCR Polymerase chain reaction
PDR Pandrug resistance
PEL Probable effect level
PFGE Pulsed-Field Gel Electrophoresis
XVIII
POP Persistent organic pollutants ppb part per billion ppm part per million ppt part per trillion
Psd Pseudomonas species
SDG Sustainable Development Goals
SQG Sediment quality guidelines
STEC Shiga-toxin producing E. coli
SVS Swiss Veterinary Society
UNEP United Nation Environment Programme
UNICEF United Nations Children’s Fund
UV Ultra-violet
VF Virulence factors
WHO World Health Organization
WGA Within Group Analysis
WWTP Municipal wastewater treatment plant
XIX
Figure caption
Figure 1-1 Scheme of a conventional WWTP with activated sludge for the removal of biodegradable organic matter, nitrification, denitrification and chemical phosphorus
removal by FeCl3 precipitation ...... 8 Figure 1-2 Combination of primary, secondary and tertiary treatment processes ...... 9 Figure 1-3 Number of unique enzymes targeting antibiotic drugs ...... 11 Figure 1-4 Major antibiotic antibiotic target in the cell: Antibiotic target vital processes in the cell by competition against the natural susbtrate, alteration of cell membranes or inhibition of transcription and traduction processes (DNA or protein synthesis)...... 13 Figure 1-5 Common β-lactam derivatives ...... 14 Figure 1-6 Mechanism of action of b-lactams antibiotics. At the Top: in absence of β-lactam antibiotic PBPs binds terminal D-Ala-D-Ala of the pentapeptide sequence of the NAM to form the secondary structure of the cell wall. At the bottom: in the presence of a β- lactam antibiotic, the β-lactam bind the PBPs avoiding the binding of the D-Ala-D-Ala. The transpeptidase activity is inhibited and the synthesis of secondary structure of the cell wall is blocked ...... 15 Figure 1-7 Principal mechanisms of β-lactam resistance. In Gram-positive bacteria, the peptidoglycan is tick (20-80nm) and the permeability low thus avoid antibiotic penetration and PBPs access. In Gram-negative bacteria, access to PBPs in easier because of the thin layer of peptidoglycan. By acting on the porins permeability, efflux system expression and enzymatic activation, bacteria can avoid antibiotic access to PBPs...... 16 Figure 1-8 B-lactam ring hydrolysis by serine β-lactamases (Class A, C and D) and by metallo-β-lactamases (Class B) ...... 17 Figure 1-9 Mechanisms of HGT in bacteria. Antibiotic resistance can be acquired by (i) Conjugation: recombination/ exchange of plasmids; (ii) Natural transformation: uptake of environmental DNA or (iii) Transduction: infection by bacteriophages...... 19 Figure 1-10 Major ICEs structures (Source : Merlin (1999)). Black triangles represent terminal inverted repetitive sequences (IR) Inserting sequences (IS) carried only a gene coding for a transposase (tnpA). Composite transposons are 2 IS containing a gene of interest XX. Non-composite transposons code for a transposase (tnpA) and a resolvase (tnpR). Red box indicate active site for the resolvase(res) and black triangles, the active sites of the transposase. Integrons are constituted by an int gene coding for an integrase,
XXI
the active site of the integrase (attI) and a promotor (Pant). The integrase can trap and mobilize one or more gene cassette downstream the promotor...... 20 Figure 1-11 Timeline of antibiotic and antibiotic resistance development ...... 22 Figure 1-12 Worldwide distribution of NDM-positive strains of the Enterobacteriaceae. In red: prevalence of NDM-positive strains among Enterobacteriaceae of ≥ 5%. In brown, prevalence of NDM-positive strains among Enterobacteriaceae of ≤ 5%. In white, countries without report ...... 23 Figure 1-13 Routes by which ARB and ARG can cycle between human population, terrestrial environment and aquatic environment ...... 24 Figure 1-14 Access coverage of the RD Congo population to drinking water, sanitation and hygiene facilities ...... 25 Figure 1-15 Pharmacies in a street of Kinshasa (RD Congo) ...... 25 Figure 1-16 Localization of the sampling sites in the province of Kinshasa, Democratic Republic of the Congo ...... 28 Figure 1-17 Unregulated waste discharge in an urban river of Kinshasa ...... 29 Figure 1-18 Lake Brêt aerial picture (Switzerland). In red, the location of Switzerland in Europe...... 30 Figure 2-1 Location map of the study area. A: Location Map of Congo DR in Africa. B: Map of Congo DR. C: Location map of studied Rivers, R1: Makelele, R2: Kalamu, R3: Nsanga at Kinshasa, Congo DR...... 45 Figure 2-2 Score plot for principal component analysis (PCA) applied to sediment mesurement across sampling sites: (a) PCA of PCBs congeners, (b) PCA of pesticides congeners, (c) PCA of PBDEs congeners and (d) PCA of PAHs congeners ...... 58 Figure 3-1 Location map of the study area. A: Location Map of Congo DR in Africa. B: Location Map of Kinshasa City in Congo DR. C: Location map of studied River (Kokolo Canal) and shallow wells P1 and P2 at Kinshasa, Congo DR...... 75 Figure 3-2 Photos taken by John Kayembe in April 2017. A: Bathing, recreational activity in Kokolo Canal; B: Kokolo canal using for domestic purpose; C: Shallow well P1 using for domestic purpose; D: Shallow well P2 using for domestic and agricultural purpose 77 Figure 3-3 Photos taken by John Kayembe in July 2017. A: Pit latrines located in the bank of Kokolo Canal; B: Recreational activity (children playing foot) near Kokolo Canal; C: Children playing in Kokolo canal; D: Children bathing/enjoying in Kokolo Canal after playing foot ...... 77
XXII
Figure 3-4 Average of Escherichia coli and Enterococcus quantification in water and sediment samples from river (Kokolo Canal), Hospital Outlet Pipe and shallow wells (P1 and P2) during the wet and dry seasons* ...... 82 Figure 4-1 Localization of the sampling site in the province of Kinshasa, Republic Democratic of Congo ...... 100 Figure 4-2 Raw 16S rRNA copy number detected in hospital receiving systems (16S rRNA gene copy number g-1 of DS) at each sampling point ...... 108 Figure 4-3 Raw FIB copy number detected in hospital receiving systems at each sampling point ...... 109 Figure 4-4 Normalized FIB copy number detected in hospital receiving systems at each sampling point...... 110 Figure 4-5 Normalized ARGs copy number detected in hospital receiving systems at each sampling point...... 110 Figure 4-6 Raw ARGs copy number detected in hospital receiving systems at each sampling point ...... 112 Figure 4-7 Grouping of each sampling point according to ARGs, FIB, toxic metals, grain size and OM. A – plot using between group analyses to discriminate each points in various hospital receiving system. B – same samples plotted after decomposing differences in each sampling point by within group analysis. Right upper panels show correlation with variables (PCA variable scores)...... 115 Figure 5-1 Normalized ARGs copy number detected in hospital receiving systems at each sampling point ...... 137 Figure 5-2 ARGs copy number g-1 DS detected in hospital receiving systems at each sampling point...... 137 Figure 5-3 Correlation circle for the studied parameters. Principle component is the x axis, with the second component being the y axis...... 138 Figure 5-4 Graph of individuals of the PCA results with correlation circles ...... 138 Figure 6-1 Localization of the sampling site in the province of Kinshasa, Republic Democratic of Congo ...... 153 Figure 6-2 Percentages of 150 ESBL-producing E. coli isolates exhibiting antimicrobial resistance...... 159 Figure 6-3 Prevalence of resistance to antibiotic classes in ESBLEC isolates. ESBLEC = 150 ...... 160
XXIII
Figure 6-4 Comparison of the distribution (%) of extended-spectrum β-lactamases (ESBLs) between sampling location ...... 161 Figure 7-1 Sampling site, Lake Brêt in Switzerland. B5-B19 sampling sites...... 175 Figure 7-2 Raw 16S rRNA copy number detected in soils and sediments (16S rRNA gene copy number g-1 DW)...... 181 Figure 7-3 Raw β-lactams resistance genes copy number detected in soils and sediments. .. 182 Figure 7-4 Normalized β-lactams resistance genes copy number detected in in soils and sediments...... 182 Figure 7-5 Raw sulfonamides resistance genes copy number detected in soils and sediments...... 183 Figure 7-6 Normalized sulfonamide resistance genes copy number detected in in soils and sediments...... 183 Figure 7-7 Correlation graphic between all parameters. Color gradient represent the R coefficient. Crossmark signifies unsignificant correlation between the parameters...... 186
XXIV
Table caption
Table 1-1 Mechanisms of β-lactam resistance influencing virulence processes ...... 21 Table 2-1 GPS location of sampling sites and physico-chemical parameters of surface sediments from Makelele, Kalamu and Nsanga Rivers ...... 50 Table 2-2 Metal content of surface sediment samples from River Makelele (R1), Kalamu (R2) and Nsanga (R3) analyzed by ICP-MS ...... 52 Table 2-3 Igeo and EF values for Cu, Zn and Pb in surface sediments...... 53 Table 2-4 Concentration (in µg kg-1 dry weight) of polychlorinated biphenyl (PCBs), and polycyclic aromatic hydrocarbons (PAHs) in sediment samples from River Makelele (R1), Kalamu (R2) and Nsanga (R3)...... 55 Table 2-5 Concentration (in µg kg-1 dry weight) of organochlorine pesticides (OCPs) and BDEs in sediment samples from river Makelele , Kalamu and Nsanga ...... 56 Table 2-6 The values of Fluo/ (Fluo + Pyr), IDP/ (IDP + BghiP), BaA/ (BaA + Chry), and LMW/HMW ratios...... 62 Table 2-7 Spearman’s rank-order correlation of selected parametersa analyzed in the surface sediments ...... 64 Table 3-1 GPS Location and description of sampling sites ...... 76 Table 3-2 Primers used for PCR amplification of general E. coli and Enterococci, and human- specific Bacteroides* ...... 79 Table 3-3 Physicochemical parameters of water samples from Kokolo Canal and wells (P1 and P2) during the dry season (dry) and wet season (wet) ...... 80 Table 3-4 Average of Escherichia coli and Enterococcus quantification in water and sediment samples from river (Kokolo Canal) and shallow wells (drinking water) during the wet and dry seasons ...... 84 Table 3-5 PCR presence/absence assays for detection of E.coli and Enterococcus in water samples from wells, rivers, hospital outlet pipes, and sediment samples from rivers (Kokolo Canal)...... 86 Table 3-6 Spearman's Rank-Order Correlation of selected parameters* analysed in water samples from Kokolo Canal...... 87 Table 4-1 Primers used in this study ...... 102 Table 4-2 Physico-chemical parameters of surface sediments from sampling points of each hospital site ...... 105
XXV
Table 4-3 Metal content of surface sediment samples from sampling point of each hospital site, analyzed by ICP-MS (expressed in mg. kg-1) ...... 106 Table 4-4 Pearson correlation ...... 114 Table 5-1 Primer table for qPCR quantitation ...... 131 Table 5-2 Physicochemical parameters in sediment samples ...... 132 Table 5-3 Metal content analyzed by ICP-MS (mg kg-1 dry weight) in sediments according to the rain event, the sampled river and the sampling location ...... 133 Table 5-4 Igeo values in surface sediments ...... 134 Table 5-5 EF values in surface sediments ...... 135 Table 5-6 PI values in surface sediments ...... 136 Table 5-7 Spearman correlation ...... 139 Table 6-1 Oligonucleotid primers used for the detection of phylogenetic groups, virulence factors, pathogenic islands and antibiotic resistance genes in strains isolated from urban rivers ...... 155 Table 6-2 Prevalence of phylogroups according to the sampling site ...... 158 Table 6-3 Phylogroup typing, antibiotic resistance, virulence factor and pathogenic island pattern of DEC strains according to their sampling location ...... 163 Table 7-1 Physicochemical parameters of sediment and soil analyzed; including Swiss coordinates, depth or altitude, total organic matter (OM) content, the proportion of clay, silt and sand, and median grain size ...... 176 Table 7-2 Metal content (mg.kg-1 dry weight) of samples according to their sampling location analyzed by ICP-MS ...... 180
XXVI
CHAPTER 1
Introduction
1
Water is a natural resource necessary for the survival of all organisms on earth including humans. With its estimation of 1’400’000’000 km3 of water, 70% of the Earth is covered by water. However, freshwater represents only 2.5% of the total water on Earth (U.S. Geological Survey). Because most of the freshwater is stored as glaciers or deep groundwater, less than 1% of the total water in lake, rivers, streams and groundwater is easily available for human use (Oki and Kanae 2006). Despite that the water circulates in a closed hydrologic cycle, the water scarcity increases across the globe because of its uneven distribution (Oki and Kanae 2006). After decade of debates, the United Nations General Assembly recognized the Human Right to Water and Sanitation through Resolution 64/292 on 28th July 2010, and with the General Comment No.15 states that “The right to water is indispensable for leading a life in human dignity. It is a prerequisite for the realization of other human rights”. Thus, the right to water was defined as the right of everyone to sufficient, safe, acceptable, physically accessible and affordable water for personal and domestic uses (UN Water 2014).
In addition, sufficient and safe water is also essential for ecosystems. Biodiversity is tightly linked to freshwater systems and freshwater organisms react to physicochemical disturbances by changing their location or by disappearing (Dudgeon et al. 2006, WMO 2013). Freshwater biodiversity plays a crucial role in many bio-geochemical processes and vice versa (Dudgeon et al. 2006). For example, phytoplankton are primary producers for the food chain, bottom dwelling species and worms affect physical-processes, many freshwater organisms sustain the water quality by acting on decomposition processes and contaminant removal (Dudgeon et al. 2006). Due to the functional redundancy of organisms, and the relationship between organisms, the precise impact of biodiversity change on the functioning of ecosystems is difficult to evaluate, however the biodiversity provides a broad variety of valuable goods and services for mankind and some of which are irreplaceable such as its direct contribution to economic productivity (fisheries) or its contribution to ecosystem services (cleaning water) (Covich et al. 2004).
The world’s freshwater services are so important that it is subject to severe competition among human stakeholders. Food, building material, prevention of flood and erosion are directly linked to freshwater resources and when water supplies are limited, conflicts can arise up to armed conflicts (Poff et al. 2003). Despite the importance of freshwater services for the humanity, degradation of water quality is almost invariably the results of human activity. The Sustainable Development Goal #6 purpose the monitoring as an essential tool for the management of water and sanitation (UN Water) in the scope that each people have the right to
2 water. The water quality monitoring provides an understanding of: (i) water-quality in aquatic systems; (ii) local, regional and national variability; (iii) temporal trends and (iv) natural and anthropogenic impact.
1.1 Anthropogenic impact on aquatic systems
Natural phenomena caused by climatic, geographical and geological condition can affect water quality by determining major ions distribution, the turbidity of rivers or other oxidation/reduction processes (WMO 2013). However, the degradation of water quality is invariably the result of anthropogenic activities.
1.1.1 Contamination sources Sources of water contamination can be divided into two categories based on their origins: point sources and non-point sources (UNEP/WHO 1996).
□ Point sources of pollution correspond to a specific location where the pollution resulted from human population and human activities. For example, it can originate from solid waste disposal, wastewater treatment plant effluents, and industrial wastewater outlet. The pollutants are discharged to a waterway through a discrete conveyance such as sewer channels and outlet pipes and the effect on the water body will be dependent of pollutant load and the capacity of the water body to dilute the discharge. The point source pollution can be identified and the polluting material can be collected and treated. □ Non-point sources of pollution correspond to diffuse or multiple-point sources that cannot be traced to a single point of discharge. Examples of diffuse pollution are agricultural and urban run-off, which enter in surface water and infiltrate into ground waters. Multiple-point sources can be defined as a multiple entry of point source pollution such as open defecation and septic tanks, which enable the monitoring of each source individually. The management of non-point sources of pollution is complex and can be only controlled by prevention.
Discharge of contaminated effluents in aquatic systems is the basic strategy for the purification of domestic effluents because of the natural capacity of the ecosystem to sustain
3 the water quality by biological processes. However, when the ecosystem can no longer cope with the flow of imported contaminant, the pollution rise.
1.1.2 Different types of pollutants Anthropogenic pollution can take many forms and have many effects. However, it can be commonly classified in two major groups: macro-pollutants and micro-pollutant.
1.1.2.1 Macro-pollutants Macro-pollutants are chemical compounds present in concentration on the order of parts per million (mg.L-1), either organic or inorganic, which can be naturally present in the environment and reach the state of pollutant when their concentration exceeds the acceptance threshold of the receiving system. The list of macro-pollutant agents is relatively short, it include acids, salts, nutrients and organic matter. Excessive nutrient input in aquatic ecosystem due to human activities usually lead to eutrophication and oxygen depletion processes (Barletta et al. 2019). Nitrogen and phosphorus are the most common pollutant of aquatic systems. In normal concentrations, N and P limits the primary production. When N and P concentrations rise, the primary production increase (plants and algae), resulting in eutrophication of aquatic systems. The proliferation of primary producers induce an absorption of large oxygen quantities leading to an asphyxia of the aquatic system and can cause the death of many species in the food chain. In addition to the impact on biodiversity, eutrophication process have secondary effects on human life style. As example, the bloom of toxic algae linked to eutrophication processes will have impact on the consummation of freshwater goods (fish, see food) and water quality (Heisler et al. 2008).
Fecal pollution of aquatic system is linked to the absence of adequate sewage system and makes water unsafe for human consumption and recreational activities. Leakage of septic tanks, open defecation or domestic wastewater insufficiently treated ends in adjacent aquatic system leading to an increase of associated bacteria in the water. The presence of fecal indicator bacteria is used as an indicator that a potential health risk exist for population exposed to this water. Indeed, many other waterborne pathogens (i.e. bacteria, virus and protozoa) are linked to fecal contamination such as cholera, enteropathogenic Escherichia coli, thyphoïdis, amoebiasis, polyomaviruses and hepatitis (Gall et al. 2015).
4
1.1.2.2 Micro-pollutants Micro-pollutant are a diverse group of chemicals. They include a much larger inventory of chemical compounds including inorganic (metals) as well as synthetic organic compound (i.e. pharmaceuticals, detergent, pesticides) that have deleterious effects at lower concentrations (ppb, ppt). They also vary in their reactivity, mode of toxicity, persistence in the environment and their potential bioaccumulation in the food chain. These compounds can have tremendous adverse effects such as feminization of fish and mussels and intersex and reproductive disruption in fish by estrogenic endocrine disruptors (Vajda et al. 2008, Margot et al. 2015).
Metals are elements that are not biodegradable and tend to accumulate in organisms. Some of them such as iron, copper and zinc are essential for life at trace levels, taking part in physiological functions of living organisms and regulate numerous physiological functions. However, an excessive level can also lead to various and serious damages in the organism (Fu and Wang 2011). Once in the environment, metal distribution will be affected by physico- chemical and biological processes leading to the dissolution of metals, its adsorption or desorption of sediment and suspended particles such as organic matter or it sedimentation and sinking in the sediment (Barletta et al. 2019). Contamination of aquatic system by metal is linked to the contamination of aquatic fauna which can bio-accumulated and even biomagnified according to the trophic level of organisms. In many contaminated sites, metals in fish tissues are above permissible levels for human consumption, thus creating a potential risk for human health. The classic example of metal poisoning is the mercury. Mercury is a neurotoxin which damage central nervous system and cause failure of pulmonary and kidney functions, chest pain and dyspnea (Namasivayam and Kadirvelu 1999). Its methylated form was the cause of the well known mass poisoning of Minamata bay responsible of 12’617 cases (2’264 cases officially recognized, 10’353 people received compensation from the government) (Minamata Disease Municipal Museum 1994). Studies of metals impact on ecosystem level are limited. However, it has been shown that metal accumulation in ecosystem is able to alter the trophic chain and modify nutrient and energy cycles (Tovar-Sánchez et al. 2018). It has been reported that metal content can exert a strong selective pressure and may be one of the most edaphic factor determining community composition (Roccotiello et al. 2015). Metal contamination can alter all the ecosystem dynamic by its impact on the distribution and abundance of population and on the community structure (Tovar-Sánchez et al. 2018). Bacterial community is among the most sensitive to metal contamination and is often used as a biomarker of pollution severity. Metatranscriptomic analysis on soils affected by metal contamination have revealed a decrease
5 in functional diversity linked to an altered microbial community composition (Jacquiod et al. 2018). Functional diversity is a key issue in ecosystem functioning, ecosystem functions are usually sustained by few but abundant species. Therefore the loss of dominant species may have strong ecosystem impact (Pan et al. 2016, Kardol et al. 2018).
Persistent organic micro-pollutants (POPs) are carbon based organic chemical compounds. POPs are typically hydrophobic and lipophilic and can be distinguished from other chemical substances by their persistence, their toxicity and their ability to be transported along large distances (Jones and de Voogt 1999, Barletta et al. 2019). There are many thousands of POPs usually described in “families” of chemicals. Among important classes of POPs are families of chlorinated aromatics, including polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and different organochlorine pesticides (i.e. DDT and its metabolites), which were initially synthetized for industrial uses or as agrochemicals. Polynuclear aromatic hydrocarbons (PAHs) are a sub-category of aromatic hydrocarbures originating mainly from an incomplete combustion process of organic matter. POPs are mostly man-made chemical products originated from numerous sources (e.g. intensive agriculture, city dumps and landfills, waste from pesticides production facilities and electrical power plants, waste incineration, road and railroads, water and wastewater treatment plants) (Bănăduc et al. 2016). In aquatic ecosystems, POPs are adsorbed on sediment or suspended solid particles in the water column. POPs may then enter in the food chain by the aquatic species diet (Bănăduc et al. 2016). Readily absorbed in fatty tissue, POPs are able to bioaccumulate and magnify up to 70’000 times in the food chain. Accumulation of POPs in tissues lead to serious adverse effects including cancer, allergies, hypersensitivity, damage to nervous system, reproductive disorders and disruption of the immune system (Jones and de Voogt 1999, Bănăduc et al. 2016). Studies on fish determined that fertility loss was linked to endocrine disruptor; that fills, kidney and liver abnormalities were linked to organochlorinated pesticides and cerebral tissues abnormalities were linked to POPs accumulation (Bănăduc et al. 2016). In term of communities’ structure and functionality, studies highlight the impact of POPs on species abundance and diversity (Bănăduc et al. 2016). In 2001, the Convention of Stockholm was adopted in order to reduce the intake of POPs in the environment. Actually, 152 countries already signed the convention (Stockholm Convention).
1.1.2.3 Pollutant of emerging concern Pollutant of emerging concern are defined as pollutant that are currently not included in routine monitoring programs and which can be candidate for future regulation, depending on their (eco)toxicity, potential health effects and public perceptions (Dulio et al. 2018). These
6 pollutants are not necessarily new, but they have been recently “discovered” by the improvement of techniques and their environmental and health significance are usually under evaluated. Among the most frequently discussed pollutant of emerging concern, we can cite industrial additive (bisphenol A, phtalates, …), pharmaceuticals (antibiotics, paracetamol, ibuprofen, benzodiazepines…), personal care products (parabens, triclosan, sun-screen agents…), steroids and hormones (androgens, estrogens, phytoestrogens…) and their metabolites (Dulio et al. 2018), nanomaterials and microplastics. Nanomaterials are materials having dimensions less than 100 nm. Nanomaterials originates from various sources including electronics, optics and medical devices. Due to their small size, nanoparticles are extremely active and possess exceptional chemical, biological and physical properties. Furthermore nanoparticles able to persist in air, water and can be easily transported (Srikanth 2019). Studies revealed that nanomaterials cause acute toxicity in multiple aquatic species especially filter- feeding invertebrates. As example of toxicity impact on aquatic organisms, nanoparticles are able to interact with cellular receptors and transport proteins resulting in ionoregulatory and respiration failure (Hyseni 2016). Microplastics, on the other hand, are residues of plastic degradation exhibiting a size less than 5 mm (Srikanth 2019). Because of their size, microplastics are freely available for ingestion to a large number of aquatic organisms. Ecotoxicological effects of microplastics are wide because of the large diversity of microplastic composition. However, studies documented the effect of neurotoxicity on fish, reduction of predatory performance in P. microps, impact on growth and reproduction on freshwater crustacean and even impact on immune response, oxidative stress and genotoxicity on molluscs (de Sá et al. 2018). Following a call by the EU Commission, the NORMAN Association was created to improve the exchange of information on emerging substances and to harmonize protocols and data quality (NORMAN 2005)
1.1.3 Contamination variability The dissemination of pollutant is subject to spatial and temporal variations. In surface water (lake, streams, wetlands, reservoirs and rivers), the hydrodynamics of the water body determined the spatial variability of pollutant deposition in all three dimensions according to the flow direction, the discharge and the time (WMO 2013). A temporal variation of the water quality can be described by determining settling rates, biodegradation rates or transport rates as well as concentrations of water parameters. Five major types of temporal variation where described by Rees and Bartram (2002): (i) Minute-to-minute to day-to-day variability, resulting 7 from water mixing and fluctuations of inputs; (ii) diurnal variability, linked to biological cycles and cycles in biological inputs; (iii) days-to-months variability, linked to climatic factors and pollution sources; (iv) seasonal hydrological and biological cycles and (v) year-to-year trends, mostly due to humans.
1.1.4 Wastewater management systems To diminish the impact of anthropogenic activities on the aquatic systems and thus its tremendous effects on aquatic fauna and flora, conventional wastewater treatment plants (WWTPs) were designed to remove the solid waste, suspended solids, easily biodegradable organic matter and nutrient (nitrogen and phosphorus) (Figure 1-1). In a preliminary treatment, coarse screening and grid remove coarse solids and other large materials. The primary treatment able to remove suspended organic and inorganic solids by sedimentation and floating materials (scum) by skimming. After the primary treatment, 25 to 50% of the incoming biochemical oxygen demand (BOD5), 50 to 70% of the total suspended solids, and 65% of the oil and grease were removed. Furthermore, a part of organic nitrogen, organic phosphorus and metals associated with solids are also removed. A secondary treatment is done on the effluent to remove residual organics by aerobic and anaerobic biological treatment processes. With the combination of primary and secondary treatment processes, 85% of BOD5 and suspended solids present in the raw wastewater as well as some heavy metals are removed.
Figure 1-1 Scheme of a conventional WWTP with activated sludge for the removal of biodegradable organic matter, nitrification, denitrification and chemical phosphorus removal by FeCl3 precipitation (Source: Margot et al. 2015).
8
When specific wastewater constituents must be removed, such as refractory organics, toxic metals and dissolved solid, tertiary treatment can be used (Figure 1-2) (Lazarova and Bahri 2005). The easiest tertiary treatment are the disinfection of secondary effluents by lagooning, chlorination, UV disinfection or ozonation (Lazarova and Bahri 2005). The membrane bioreactor (BRM) combine secondary treatment with activated sludge with membrane filtration (instead of clarification process) leading to an almost total disinfection (Lazarova and Bahri 2005).
Figure 1-2 Combination of primary, secondary and tertiary treatment processes (Source: Suez Environment)
The fate of micro-pollutant during wastewater treatment depends mainly on their physico-chemical characteristics (hydrophobicity, volatility and biodegradability) and the type of treatment. In the case of pharmaceuticals, removal efficiency may vary from 0 to 100%, depending on the compound. Some analgesics / anti-inflammatory drugs as well as natural hormones are well removed during the treatment process, whereas most of them are only partially or not removed at all (Margot et al. 2015). Removal of pesticides/ biocides in WWTPs is highly variable, but the average removal is less than 50% (Margot et al. 2015). In the case of POPs, an average removal of 50-80% for POPs pesticides and 75% for non-pesticides POPs such as PCBs was observed (Margot et al. 2015).
9
In many developing countries, the treatment of wastewater is challenging and strongly depending on the availability of adequate infrastructure facilities. On-site sanitation system such as pit latrines and septic tanks are widely used. However, the maintenance of these pit latrine and septic tanks is very poor and their leakage affects adjacent aquatic systems and groundwater (UNEP 2010). Off-site wastewater treatment are also a common practice in urban areas. Wastewater is collected and transported to WWTPs for treatment (Mara and Alabaster 2008). However, in absence of sanitation facilities, population are practicing open defecation and reject their wastewater directly in adjacent aquatic systems (UNEP 2010).
1.2 Antibiotic resistance genes as emerging pollutant
Since the first introduction of the sulfonamides in 1937, mechanisms of antimicrobial resistance evolved to reach thousands of genes (Lakin et al. 2017). Most of them target β- lactams (Figure 1-3). Facing this evolution, a call-to-arms was launched by various groups such as United Nations, FAO, WHO, CDC, and the European Commission, to increase the attention to antimicrobial resistance at the international level (Lakin et al. 2017). After the release of the global report on antibiotic resistance surveillance in 2014 and its call to improve and coordinate a global ARB surveillance (WHO 2014), the network for the monitoring of emerging environmental substances launched a call for a screening campaign of selected antibiotic resistance determinant and mobile genetics elements (AR/MGE) from WWTPs in Europe in 2015 (NORMAN). Antimicrobial resistance genes became emerging pollutant.
10
Figure 1-3 Number of unique enzymes targeting antibiotic drugs (Source: adapted from https://megares.meglab.org/, accessed 14/04/2019)
1.2.1 A brief overview of antibiotic resistance It is quite hard to determine a single definition of antibiotic resistance. For a clinician, a bacterial strain is resistant to an antibiotic when the treatment is not effective; for a pharmacologist, a bacterial strain is resistant to an antibiotic if the concentrations reached at the active site are below the minimum inhibitory concentration; for a microbiologist, a bacterial strain is resistant to an antibiotic if it has a mechanism of resistance increasing the value of the minimum inhibitory concentration; for the epidemiologist, a bacterium is resistant to an antibiotic if it has a minimum inhibitory concentration significantly different from that of the normal population (AFSSA 2006).
The environmental point of view is closer from the sentence of Courvalin and Leclercq (2006): “Resistance is not a cyclical or temporary phenomenon but a property intrinsic of the bacterial world”. On an evolutionary scale, the exponential increase of antibiotic-resistant phenotype is very recent. Resistance elements were detected in permafrost sediment from the Late Pleistocene in the Yukon and evaluated at about 30’000 calendar years (D'Costa et al. 2011). Phylogenetic analysis dates the origin of serine β-lactamase at over 2 billion years ago and suggests that many of these enzymes were plasmid encoded for million years (Hall and
11
Barlow 2004). This finding shows the existence of antimicrobial resistance largely before the antibiotic era and points towards an ecological significance of bacterial resistance. Antibiotic production is a frequent characteristic of environmental microorganisms, particularly in Actinomycete genus. Antibiotic able producers to outcompete with other bacteria and the production of antibiotic induce the necessity of self-protection such as an endogenous system of detoxification. It has been suggested that instead of being a weapon for bacteria, antibiotics may be also signaling molecules mediating intercellular communication in natural environment (Davies et al. 2006) and could play a metabolic role that may include biosynthesis of macromolecules, maintenance of homeostasis, and signal trafficking by tacking part to the quorum-sensing (Aminov 2009, Fajardo et al. 2009). The more classic role of “weapon” may only occur in some local micro-environments where high antibiotic concentration were produced (Fajardo et al. 2009). This ecological significance of antibiotic resistance is overrated by human. The selection pressure caused by the ever-increasing use of antibiotics fastens the evolution of resistance genes, selects resistance phenotypes and induces the horizontal gene transfer. Less than one hundred years after the introduction of the first antibiotic in human medicine leading to the golden age of antibiotics and the treatment of many untreated diseases, the ecological significance of antimicrobial resistance has disappeared in favor of the anthropocentric viewpoint and is actually defined as a threat “as important as climate change for the world” (Davies 2013).
1.2.2 General mechanisms of antibiotic resistance Several classes of antibiotics are available. They target vital microbial biochemistry such as translation, DNA replication, and cell wall synthesis (Figure 1-4). Resistance to different antibiotic classes is the result of altered molecular target, efflux of antibiotics out the cell, chemical modification or destruction of the antibiotic and blockade of antibiotic entry into the cell.
12
Figure 1-4 Major antibiotic target in the cell (Source: Grenni 2018): Antibiotic target vital processes in the cell by competition against the natural substrate, alteration of cell membranes or inhibition of transcription and translation processes (DNA or protein synthesis).
The resistance phenomenon can by intrinsic or acquired by the bacterium. Intrinsic resistance is a characteristic of a bacterial species shared by all the strains in this species, as a result of inherent structural or functional characteristics, in order to resist against the action of a specific antibiotic (Blair et al. 2014). For example, the lipopeptide daptomycin is not effective against Gram-negative bacteria due to the composition of the cytoplasmic membrane. Gram- negative bacteria has a lower proportion of anionic phospholipids in the cytoplasmic membrane than Gram-positive bacteria, thus resulting in a lower efficiency of the Ca2+-mediated insertion of daptomycin in the cytoplasmic membrane (Randall et al. 2013). In the case of Mycoplasma pneumoniae, the bacteria do not have a cell wall thus preventing the action of β-lactams (Cavallo et al. 2004). Acquired resistance is the result of a genetic modification by mutation in chromosomal genes or by horizontal gene transfer. Acquired resistance mechanisms are generally divided into three groups: (i) minimizing the intracellular concentration of the antibiotic by the lower penetration in the bacterium or by an efflux pump; (ii) modification of the target by genetic mutation or post-transcriptional modification; and (iii) inactivation of the
13 antibiotic by hydrolysis or modification. The acquisition result in a new resistance spectra in only some strains of the species (Blair et al. 2014).
1.2.3 Β-lactams and β-lactam resistance Β-lactams antibiotics are a wide family, divided into 3 groups according to their structure: the derivative from 6-amino-penicillanic acid, the derivative from 7-amino- cephalosporanic acid and monobactams (Figure 1-5).
Figure 1-5 Common β-lactam derivatives (Source: adapted from Cavallo 2004)
B-lactams are suicides substrates, sterically similar to the terminal D-Ala-D-Ala sequence of the N-acetylmuramic acid (NAM) pentapeptide that serve as a natural substrate for the cell wall transpeptidase. When Penicillin Binding Proteins (PBPs, i.e. transpeptidases and carboxypeptidases) recognize the substrate structure, PBPs active site fixed themselves to the β-lactam ring resulting in the inactivation of the active site by the formation of a covalent complex antibiotic-enzyme. The inactivation of PBPs lead to the irreversible inhibition of peptidoglycan synthesis (Figure 1-6), thus to the fragility of the structure and cell lysis (Cavallo et al. 2004).
14
Figure 1-6 Mechanism of action of β-lactams antibiotics. At the Top: in absence of β-lactam antibiotic PBPs binds terminal D-Ala-D-Ala of the pentapeptide sequence of the NAM to form the secondary structure of the cell wall. At the bottom: in the presence of a β-lactam antibiotic, the β-lactam bind the PBPs avoiding the binding of the D-Ala-D-Ala. The transpeptidase activity is inhibited and the synthesis of secondary structure of the cell wall is blocked (Source: http://tmedweb.tulane.edu/pharmwiki/). ABX : antibiotic; PBP: penicillin binding protein; GT glycosyl transferase, NAG: N-acetylglucosamic acid; NAM: N-acetylmuramic acid
Acquired resistance to β-lactams antibiotics is tightly linked to the necessity that the antibiotic need to reach the PBPs active site. Four resistance mechanisms are identified (Cavallo et al. 2004) as described in Figure 1-7:
- Decrease of porins permeability: β-lactams are hydrophilic molecules that need the help of porins to cross the cell wall. Structural modification or quantitative diminution of porins tend to avoid/ limit the penetration of the antibiotic in the periplasmic resulting to a resistance phenotype. - Mutation on active efflux systems who lead to a hyperexpression of the efflux system.
15
- Modification of PBPs by mutation resulting in a lower affinity for β-lactams antibiotics or the use of alternative transpeptidases for peptidoglycan synthesis - Production of a β-lactamase in high concentrations obtained by HGT acquisition or by the mutation of a constitutive β-lactamase leading to an enzyme hyperproduction.
Figure 1-7 Principal mechanisms of β-lactam resistance (Source: Cavallo 2004). In Gram- positive bacteria, the peptidoglycan is tick (20-80nm) and the permeability low thus avoiding antibiotic penetration and PBPs access. In Gram-negative bacteria, access to PBPs is easier because of the thin layer of peptidoglycan. By acting on the porins permeability, efflux system expression and enzymatic activation, bacteria can avoid antibiotic access to PBPs.
Enzymatic production is the predominant mechanism of β-lactam resistance in Gram- negative bacteria. Β-lactamases are able to open the β-lactam ring by hydrolysis thus enabling the PBPs to recognize the covalent structure (Figure 1-8). Actually, almost 2’800 unique enzymes has been identified (Bush 2018) and classified in 4 groups according to their protein sequence [Ambler classification (Ambler 1980)]: class A, B, C (serine enzymes) et D (metalloenzyme). Functional classification of Bush and Jacoby (2010), reflect the activity spectra of the enzyme. This classification divide β-lactamases in 3 groups according to the substrate (penicillin, oxacillin, carbenicillin, 3G cephalosporin and imipenem) as well as their inhibition profile (clavulanic acid or EDTA).
16
- Group 1 cephalosporinases (class C cephalosporinases) (i.e. AmpC) - Group 2 serine β-lactamases (class A and D)(i.e. TEM, SHV, CTX, OXA) - Group 3 metallo-β-lactamases (Class B) (i.e. IMP, VIM, NDM)
Figure 1-8 B-lactam ring hydrolysis by serine β-lactamases (Class A, C and D) and by metallo-β-lactamases (Class B) (Source: Morar 2010)
1.2.4 Genetic support and transfer mechanisms of antimicrobial resistance Genes conferring antibiotic resistance in a microorganism, called antibiotic resistance genes (ARGs), can be chromosomally or extra-chromosomally encoded. Resistance material can be acquired by integration of exogenous genes (plasmids, transposons, integrons and bacteriophages), mutation of cellular genes or combination of these processes.
Spontaneous beneficial mutation of chromosomal genes are quite rare because mutation are generally deleterious for the organisms. Spontaneous mutations are usually errors of replication or incorrect repairment of damaged DNA. For example point mutation in PBPs leads to β-lactam resistance in Salmonella enterica var Typhimurium LT2 (Sun et al. 2014). However, antibiotic use can lead to a hypermutable phenotype by the induction of a SOS
17 mutagenic response in E. coli, Salmonella typhimurium, Bacillus subtilis, Pseudomonas sp. and Clostridium sp. (Foster 2000). Ecologically, this transient state may be bacterial altruism for the population survival by increasing the genetic variability of the population (Foster 2000).
Horizontal gene transfer (HGT) characterizes the transfer of resistance genes from one bacteria to another one. HGT can happen between phylogenetically distinct bacteria and even between Gram-positive and Gram-negative bacteria. Genes can be transferred by conjugation, transformation or transduction (Figure 1-9). Transformation, the first discovered mechanism, corresponds to the stable uptake, integration and functional expression of extracellular DNA. It occurs when bacteria are in a physiological state of competence in response to a specific environmental condition such as altered growth condition, nutrient access, cell density or starvation. Only a few bacterial species are able to develop a natural state of competence and it includes human pathogens (Campylobacter, Haemophilus, Helicobacter, Neisseria, Pseudomonas, Staphylococcus and Streptococcus) (Thomas and Nielsen 2005). Interestingly, some of these naturally competent bacteria are part of the WHO list pathogen (A. baumannii, P. aeruginosa, S. aureus, H. pylori, N. gonorrhoeae, Campylobacter spp., S. pneumonia and H. influenza) (WHO/EMP 2017). DNA uptake in the environment can originate from active excretion of DNA by some genera of bacteria (Acinetobacter, Alcalingenes, Azotobacter, Bacillus, Flavobacterium, Micrococcus, Pseudomonas and Streptococcus) or passive release of DNA from dead bacteria. Transduction corresponds to the delivery of genetic material through phage predation. After infection by a bacteriophage, bacterial DNA is sometimes accidentally packaged in the capsid of the bacteriophage. When the bacteriophage infects another bacteria, it can inject the foreign DNA. The foreign DNA can be recombined into the genome of the recipient cell and then expressed.
18
Figure 1-9 Mechanisms of HGT in bacteria (Source: Schroeder 2017). Antibiotic resistance can be acquired by (i) Conjugation: recombination/ exchange of plasmids; (ii) Natural transformation: uptake of environmental DNA or (iii) Transduction: infection by bacteriophages.
Conjugative transfer is mediated by cell-to-cell junction and a pore through which DNA pass. Conjugative transfer can happen by (i) the transfer of an auto-transferable plasmid, such as F- plasmid or RP4 plasmid of E. coli; (ii) mobilization of a conjugative plasmid, (iii) co-integration of 2 plasmids; or (iv) transfer of conjugative transposon (Filutowicz et al. 2008). In Gram- negative bacteria, a type IV secretion system (T4SS) called transferosome spans the cell envelope and is responsible of the synthesis of a conjugative pilus. The relaxosome (a protein complex) processes the DNA at the origin of transfer (oriT) and delivers it to the transferosome by the coupling protein (T4CP) which connects the transferosome to the relaxosome and forms the conjugative pore. At nic site in the oriT, the relaxase cleave the plasmid resulting in the transfer of a single stranded DNA (ssDNA) in a 5’-3’ direction to the recipient cell.
HGT processes largely depend on integrative and conjugative elements (ICEs) a diverse group of mobile genetic elements (MGEs), which encode enzymes and proteins mediating the movement of DNA within genomes or between bacterial cells (Frost et al. 2005). Primarily, ICEs used to be in the host chromosome. However, ICEs have the ability to excise and transfer
19 by conjugation by a large range of core function of integration, excision, transfer and regulation (Merlin and Toussaint 1999). Majors ICEs are insertion sequences, transposons and integrons (Figure 1-10). Insertion sequence (IS) are the simplest transposable elements, which only encode necessary enzymes for their own transposition (tnpA). Composite transposons are bigger, they are composed by two IS close enough which can mobilize the DNA segment they framed (noted XX in the Fig. 1-10). The DNA fragment between the two IS can code for any functions such as antibiotic resistance (e.g. Tn5 and Tn10), catabolism genes (e.g. Tn3411). Non-composite transposons do not carry IS and contain all necessary genes for transposition process (a transposase TnpA and a resolvase TnpR) as well as auxiliary genes such as catabolism genes (e.g. Tn4651) or antibiotic resistance genes (e.g. Tn6). Red boxes in Fig. 1- 10 shows action sites of the resolvase res and black triangles, the action sites of transposases. Integrons are mobilization elements, they trapped DNA sequences called “cassette” downstream the promotor. Integrons are able to trap many cassettes which will be expressed by the same promotor.
Figure 1-10 Major ICEs structures (Source : Merlin 1999). Black triangles represent terminal inverted repetitive sequences (IR) Inserting sequences (IS) carried only a gene coding for a transposase (tnpA). Composite transposons are two IS containing a gene of interest XX. Non- composite transposons code for a transposase (tnpA) and a resolvase (tnpR). Red box indicate active site for the resolvase(res) and black triangles, the active sites of the transposase. Integrons are constituted by an int gene coding for an integrase, the active site of the integrase (attI) and a promotor (Pant). The integrase can trap and mobilize one or more gene cassette downstream the promotor.
20
1.2.5 Link between AMR and virulence Transmission of antibiotic resistance and virulence share common mechanisms. Virulence factors (VFs) are generally associated with MGEs. For example, the pathogenicity of the Shiga-toxin producing E. coli (STEC) is associated with the presence of the large virulence plasmid pO157, which can be horizontally transferred (Schroeder et al. 2017). Furthermore, the regulation of antimicrobial resistance and virulence are intertwined and connected. The regulation of virulence can affect antibiotic resistance gene expression and vice versa because of common form of gene regulation such as quorum sensing. For example, by increasing the biofilm formation, which is a virulence property, the penetration of antibiotic in the bacterial cell can be limited (Anderl et al. 2000). Furthermore, multidrug efflux pumps are involved in both antimicrobial resistance by regulating the antibiotic concentration in cell and in virulence-related process such as iron acquisition (Alcalde-Rico et al. 2016). In the case of β-lactam resistance, numerous publication highlighted the impact of antibiotic on virulence properties (Beceiro et al. 2013). Β-lactam resistance linked to PBPs modifications, production of β-lactamases, porins and efflux pumps modifications can also affect virulence properties, mainly by affecting the fitness cost, invasion process and the biofilm formation (Table 1-1).
Table 1-1 Mechanisms of β-lactam resistance influencing virulence processes (Source: Beceiro 2013)
Mechanism of resistance Implication in virulence PBP modification PBP2 (mecA) Regulation of Agr quorum-sensing system; biofilm formation; attenuated virulence in mouse model; infection persistence SCCmec Expression of phenol-soluble modulins PBP2a-PBPX Attenuated virulence in mouse model PBP7-8 Attenuated virulence in mouse model Β-lactamases CTX-M-type ESBLs Usually plasmid born; increased virulence non clearly demonstrated OXA-10-like, OXA24, Fitness cost in common host (changes in peptidoglycan composition) and SFO-1 AmpC Fitness cost AmpC/AmpD/AmpR Fitness cost and virulence; AmpR (transcriptional regulator of ampC) also controls expression of alginate production and quorum-sensing system; type 3 fimbrial gene expression and biofilm formation Porins OmpA Adhesion cell; induction of cell death; biofilm formation OprD-like Attenuated virulence in mouse model Efflux pumps AdeABC Colonization, infection, and persistence of microorganism in host AcrAB-TolC Colonization, infection, and persistence of microorganism in host
21
1.2.6 The global spread of antibiotic resistance and the rise of superbugs Antibiotic resistance has been existing for billion years (Hall and Barlow 2004). However, the massive production and use of antibiotics since the 1940s lead to an exponential increase of antimicrobial resistance. Each time an antimicrobial was introduced, a global resistance was detected (Figure 1-11). In 1955, the first multidrug resistant (MDR) Shigella was detected, whereas MDR strains were still rare. Few years later, 10% of Shigella strains were MDR (Chopra and Roberts 2001). The first recognition of an Extending Spectra β-lactamase (ESBL) from the CTX-M β-lactamase occurred in 1989 at Munich (Bauernfeind et al. 1990). It reported a cefotaxim-resistant but ceftazidime susceptible E. coli. In 1990, a cefotaxim resistance Salmonella typhimurium was recovered in South America (Bauernfeind et al. 1992). A later sequencing confirm 84% of homology between the two CTX-M genes (Bauernfeind et al. 1996). Actually, CTX-M enzymes are also present in wild life and environmental compartment. Furthermore, CTX-M gene emerges in other Enterobacteriaceae than E. coli, K. pneumoniae and Salmonella spp.
Figure 1-11 Timeline of antibiotic and antibiotic resistance development (Source: Ventola 2015)
22
The most serious threat remains the dissemination of superbugs, such as Carbapenem- resistant Enterobacteriaceae (CRE). Superbugs are commensal and pathogenic bacteria carrying multidrug resistance genes. The global spread of superbugs is well illustrated by the New Delhi metallo-β-lactamase, which was firstly isolated from Klebsiella pneumonia and Escherichia coli in Sweden from an Indian patient recently transferred from New Delhi hospital (Nordmann et al. 2011). A study reveal that the NDM-1 gene was already widespread in the Indian subcontinent in 2010 (Kumarasamy et al. 2010). Last data on NDM gene, highlight a global dissemination across the globe (Wu et al. 2019) as shown in Figure 1-12.
Figure 1-12 Worldwide distribution of NDM-positive strains of the Enterobacteriaceae. In red: prevalence of NDM-positive strains among Enterobacteriaceae of ≥ 5%. In brown, prevalence of NDM-positive strains among Enterobacteriaceae of ≤ 5%. In white, countries without report (Source: Wu 2019)
1.2.7 Relevance of antibiotic resistance in the environment Antibiotic resistance is not restricted to pathogenic or commensal bacteria. However, the anthropocentric point of view focuses the attention of AMR dissemination in clinical settings. Since a decade and enhanced by the one health approach, the number of studies aiming to determine the contribution of environmental compartment in AMR dissemination began to grow. Indeed, all antibiotic residues, antibiotic resistant bacteria (ARBs) and ARGs coming from human and veterinary use as well as industrial practices end in the environmental compartment due to the lack of efficiency of waste and wastewater management or by the
23 leakage of soils (Figure 1-13). Hotspot for the maintenance, the mixing and the gene exchange between environmental and pathogenic bacteria are mostly man made (such as sewage, WWTPs, animal farms) (Baquero et al. 2008) and environmental transfer of ARGs/ARBs to human was already described (Marshall and Levy 2011). A study from Cabello et al. (2017), suggests that the use of colistin in Asian aquaculture may be correlated to the emergence of plasmids encoding colistin resistance. It has been shown that resistant determinant contracted by contact in a contaminated environment can be maintained within the gut microbiota for several months (Johnson and Woodford 2013, Walsh 2013). Furthermore, a study of Blaak et al. (2014) revealed that the AMR contamination of fresh vegetable is an indicator of the global AMR contamination of their agricultural environment. Thus man made environment can be potential vector of ARBs and ARGs between hosts through the environment (Singer et al. 2016).
Figure 1-13 Routes by which ARB and ARG can cycle between human population, terrestrial environment and aquatic environment (Source: Taylor 2011)
In developing countries, poverty is the root of AMR dissemination. The urban poor often resides in overcrowded accommodation, with a global lack of access to hygiene and sanitation facilities. For example, 2015 household data in RD Congo (Wash project) highlight that more than 80% of the population have limited or no access to hygiene facilities and that more than 50% of the population used unimproved sanitation facilities (Figure 1-14). Furthermore, in relation with the lack of sanitations facilities, most of the solid and liquid wastes are generally
24 discharged into the adjacent aquatic system leading to a strong level of contamination of various macro-, micro- and emerging pollutants.
The lack of sanitation leads also to a predominant disease burden including water borne disease, thus explaining the need to drug access. However, drug use and deliverance is often poorly regulated (Figure 1-15) and health professionals usually have a lack of knowledge on drug delivery (Nzolo et al. 2013). Knowing that antibiotics are poorly metabolized and that fecal pollution is the main driver of AMR abundance in environments highly impacted by human population (Karkman et al. 2019), the environmental compartments in developing countries should be an important vector of AMR maintenance and dissemination.
Figure 1-14 Access coverage of the RD Congo population to drinking water, sanitation and hygiene facilities
Figure 1-15 Pharmacies in a street of Kinshasa (RD Congo) (Source: A. Laffite June 2016)
25
1.3 Research objectives and chapters’ organization 1.3.1 Research objectives A growing number of publications is available on micropollutants and ARGs dissemination all over the globe. However, there is a crucial lack of data concerning the case of developing countries. WHO Global report on AMR surveillance determined that the true extend on the AMR problematic in the African Region is actually limited because of the lack of available data. The main objective of the present interdisciplinary research is, therefore, to assess and to perform a comparative survey on the prevalence and dissemination of toxic metals, POPs, FIBs, ARBs and ARGs: broad-spectrum β-lactam and sulfonamide resistance genes (blaSHV, blaCTX-M, blaNDM, blaKPC, blaOXA-48, blaVIM, blaIMP, sul1 and sul2) and the total bacterial load (16S rRNA genes) in several receiving systems with respect to degree of wastewater management. Most of the work is focused on the receiving river system of the vicinity of Kinshasa (RD Congo) as a case study. For comparative purposes, the ARGs load in antibiotic resistance in Lake Brêt (Switzerland), non-impacted by human wastewaters was also explored. The specific objectives of the work were:
To discuss the occurrence and spatial distribution of toxic metals (Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, Ag, Cd, Sn, Sb, Pb and Hg), persistent organic polluants (including OCPs, PCBs, PBDEs) and PAHs in receiving systems under tropical conditions. To explore the effect of the lack of sanitation facilities on the microbial water quality of rivers and wells according to the seasonal variation.
To determine the prevalence and dissemination of ARGs (blaTEM, blaCTX-M,
blaSHV, and aadA), total bacterial load, and selected bacterial marker genes of fecal indicator bacteria (E. coli and Enterococcus) and Pseudomonas species in river receiving hospital effluents under tropical conditions. To investigate the virulence genes, β-lactamase and carbapenemase resistance genes in the river sediments receiving effluent waters and assessment of potential human and environmental impacts.
To quantify broad-spectrum β-lactam and sulfonamide resistance genes (blaTEM,
blaSHV, blaCTX-M, blaNDM, sul1 and sul2) and the total bacterial load (16S rRNA genes) from the total DNA extracted from the surface sediments of the Lake Brêt, Switzerland and make comparison with the developing country.
26
1.3.2 Thesis outline The chapters presented in this manuscript are part of an overall study, which focus on the dissemination of organic and inorganic micropollutant such as toxic metals, PCBs, HAPs, pesticides, FIBs, ARBs and ARGs. Each chapter is self-containing and represents the form of a published article or article in the process of being submitted. The introduction chapter is followed by 6 research chapters presenting the main results of the work:
Chapter 2 provides results on the quantification of toxic metals and POPs in three urban rivers of Kinshasa (RD Congo) and the assessment of the degree and the origin of anthropogenic pollution.
Chapter 3 assesses the impact of the lack of wastewater management on the microbial water quality by the determination of water physico-chemical parameters, fecal contamination in an urban river and two shallow wells in the vicinity of Kinshasa (RD Congo).
Chapter 4 introduces the problematic of AMR with the determination of the accumulation of toxic metals and ARGs in four urban rivers (Kinshasa, RD Congo) impacted by a hospital wastewater outlet pipe.
Chapter 5 focuses on Carbapenem resistance dissemination in aquatic system with the quantification of IMP, VIM, KPC, OXA-48 and NDM genes in tropical urban rivers (Kinshasa, RD Congo).
Chapter 6 presents the link between AMR and virulence in 3G Cephalosporin resistant E. coli isolated from tropical urban river.
Chapter 7 focuses on the metal and ARGs load in a small peri-alpine lake non-impacted by human wastewaters. This lake is only subjected on agriculture and breeding practice thus highlighting the differential impact of animal vs. human on AMR and metals dissemination.
1.3.3 Study site implicated in the research 1.3.3.1 Urban rivers of Kinshasa Kinshasa is the capital and largest city in the Republic Democratic of the Congo (Figure 1 17). Kinshasa is one of the biggest city in the world with around 12 million inhabitants in
27
2017 and an area of 9’965 km². The climate is tropical with an average temperature of 25°C and a mean of annual precipitation of 1274 mm.
Figure 1-16 Localization of the sampling sites in the province of Kinshasa, Democratic Republic of the Congo (Source: Laffite 2016)
With the Congo River and abundant precipitations, the country possesses 52% of water resources in Africa. However, the country suffers of chronical crisis of water supply. The rural exodus and the migrations following civil wars in the East led to an exponential increase of the population over the years and urban infrastructure did not evolve at the same speed as human population. Thus, a large part of the population suffers of bad sanitation conditions and lack of access to safe water. Additionally, no WWTP are implemented in Kinsahasa and all wastewater is directly discharged in urban rivers without any prior treatment, thus resulting in a global latent pollution in the city (Figure 1-17).
28
Figure 1-17 Unregulated waste discharge in an urban river of Kinshasa (Source: A. Laffite, June 2016)
1.3.3.2 Lake Bret Lake Brêt is a small perialpine eutrophic lake located in Western Switzerland at an elevation of 674 m a.s.l. (Figure 1-18). The lake has a surface area of 0.36 km² and a large catchment area of 23 km². Precipitation, spring water, minor tributaries and the derivation of the Grenet River aliment the lake Brêt. The lake does not have any outlet, all the water is used to aliment the water treatment plant to supply drinking water to the city of Lausanne (Lods- Crozet et al. 2009, Thevenon et al. 2013, Laffite et al. 2019).
29
Figure 1-18 Lake Brêt aerial picture (Switzerland) (Source: Google Earth). In red, the location of Switzerland in Europe.
1.3.4 Institutional framework This interdisciplinary research was developped under the regulations of the University of Geneva, Faculty of Sciences, section of Earth Sciences and Environment, Department F.-A. Forel for Environmental and Aquatic Sciences. The doctoral study was completed through the collaboration of the following institutions:
University of Geneva (UNIGE), Faculty of Sciences, section of Earth Sciences and Environment, Department F.-A. Forel for Environmental and Aquatic Sciences, Bd Carl-Vogt 66, CH-1211 Geneva University of Kinshasa (UNIKIN), Faculty of Sciences, B.P. 190, Kinshasa XI, Republic Democratic of the Congo National Pedagogic University (UPN), Quartier Binza/UPN, B.P. 8815 Kinshasa, Democratic Republic of the Congo
30
1.3.5 Funding Swiss National Science Foundation (grant n° 31003A_150163/1)
1.3.6 List of publication relative to this research Laffite, A., D. M. M. Al Salah, V. I. Slaveykova and J. Poté (2019). "Prevalence of β- Lactam and Sulfonamide Resistance Genes in a Freshwater Reservoir, Lake Brêt, Switzerland." Laffite, A., P. I. Kilunga, J. M. Kayembe, N. Devarajan, C. K. Mulaji, G. Giuliani, V. I. Slaveykova and J. Poté (2016). "Hospital Effluents Are One of Several Sources of Metal, Antibiotic Resistance Genes, and Bacterial Markers Disseminated in Sub- Saharan Urban Rivers." Frontiers in microbiology 7: 1128-1128. Atibu, E. K., N. Devarajan, A. Laffite, G. Giuliani, J. A. Salumu, R. C. Muteb, C. K. Mulaji, J.-P. Otamonga, V. Elongo, P. T. Mpiana and J. Poté (2016). "Assessment of trace metal and rare earth elements contamination in rivers around abandoned and active mine areas. The case of Lubumbashi River and Tshamilemba Canal, Katanga, Democratic Republic of the Congo." Chemie der Erde - Geochemistry 76(3): 353-362. Devarajan, N., A. Laffite, N. D. Graham, M. Meijer, K. Prabakar, J. I. Mubedi, V. Elongo, P. T. Mpiana, B. W. Ibelings, W. Wildi and J. Pote (2015). "Accumulation of clinically relevant antibiotic-resistance genes, bacterial load, and metals in freshwater lake sediments in Central Europe." Environ Sci Technol 49(11): 6528-6537. Devarajan, N., A. Laffite, C. K. Mulaji, J.-P. Otamonga, P. T. Mpiana, J. I. Mubedi, K. Prabakar, B. W. Ibelings and J. Poté (2016). "Occurrence of Antibiotic Resistance Genes and Bacterial Markers in a Tropical River Receiving Hospital and Urban Wastewaters." PLoS One 11(2): e0149211-e0149211. Devarajan, N., A. Laffite, P. Ngelikoto, V. Elongo, K. Prabakar, J. Mubedi, P. M. Piana, W. Wildi and J. Poté (2015). "Hospital and urban effluent waters as a source of accumulation of toxic metals in the sediment receiving system of the Cauvery River, Tiruchirappalli, Tamil Nadu, India." Environmental Science and Pollution Research 22(17): 12941-12950. Kapembo, M. L., D. M. M. Al Salah, F. Thevenon, A. Laffite, M. K. Bokolo, C. K. Mulaji, P. T. Mpiana and J. Poté (2019). "Prevalence of water-related diseases and groundwater (drinking-water) contamination in the suburban municipality of Mont
31
Ngafula, Kinshasa (Democratic Republic of the Congo)." Journal of Environmental Science and Health, Part A: 1-11. Kapembo, M. L., A. Laffite, M. K. Bokolo, A. L. Mbanga, M. M. Maya-Vangua, J.-P. Otamonga, C. K. Mulaji, P. T. Mpiana, W. Wildi and J. Poté (2016). "Evaluation of Water Quality from Suburban Shallow Wells Under Tropical Conditions According to the Seasonal Variation, Bumbu, Kinshasa, Democratic Republic of the Congo." Exposure & health 8(4): 487-496. Kayembe, J. M., F. Thevenon, A. Laffite, P. Sivalingam, P. Ngelinkoto, C. K. Mulaji, J.-P. Otamonga, J. I. Mubedi and J. Poté (2018). "Corrigendum to the paper: High levels of faecal contamination in drinking groundwater and recreational water due to poor sanitation, in the sub-rural neighbourhoods of Kinshasa, Democratic Republic of the Congo by Kayembe et al., (2018)." International Journal of Hygiene and Environmental Health. Kayembe, J. M., F. Thevenon, A. Laffite, P. Sivalingam, P. Ngelinkoto, C. K. Mulaji, J.-P. Otamonga, J. I. Mubedi and J. Poté (2018). "High levels of faecal contamination in drinking groundwater and recreational water due to poor sanitation, in the sub-rural neighbourhoods of Kinshasa, Democratic Republic of the Congo." International journal of hygiene and environmental health 221(3): 400-408. Kilunga, P. I., J. M. Kayembe, A. Laffite, F. Thevenon, N. Devarajan, C. K. Mulaji, J. I. Mubedi, Z. G. Yav, J.-P. Otamonga, P. T. Mpiana and J. Poté (2016). "The impact of hospital and urban wastewaters on the bacteriological contamination of the water resources in Kinshasa, Democratic Republic of Congo." Journal of environmental science and health. Part A, Toxic/hazardous substances & environmental engineering 51(12): 1034-1042. Kilunga, P. I., P. Sivalingam, A. Laffite, D. Grandjean, C. K. Mulaji, L. F. de Alencastro, P. T. Mpiana and J. Poté (2017). "Accumulation of toxic metals and organic micro-pollutants in sediments from tropical urban rivers, Kinshasa, Democratic Republic of the Congo." Chemosphere 179: 37-48. Mavakala, B. K., S. Le Faucheur, C. K. Mulaji, A. Laffite, N. Devarajan, E. M. Biey, G. Giuliani, J.-P. Otamonga, P. Kabatusuila, P. T. Mpiana and J. Pote-Wembonyama (2016). "Leachates draining from controlled municipal solid waste landfill: Detailed geochemical characterization and toxicity tests." Waste Management 55.
32
Nienie, A. B., P. Sivalingam, A. Laffite, P. Ngelinkoto, J.-P. Otamonga, A. Matand, C. K. Mulaji, E. M. Biey, P. T. Mpiana and J. Poté (2017). "Microbiological quality of water in a city with persistent and recurrent waterborne diseases under tropical sub-rural conditions: The case of Kikwit City, Democratic Republic of the Congo." International journal of hygiene and environmental health 220(5): 820-828. Nienie, A. B., P. Sivalingam, A. Laffite, P. Ngelinkoto, J.-P. Otamonga, A. Matand, C. K. Mulaji, J. I. Mubedi, P. T. Mpiana and J. Poté (2017). "Seasonal variability of water quality by physicochemical indexes and traceable metals in suburban area in Kikwit, Democratic Republic of the Congo." International soil and water conservation research 5(2): 158-165.
33
References
AFSSA (2006). Usages vétérinaires des antibiotiques, résistance bactérienne et conséquence pour la santé humaine Agence française de sécurité sanitaire des alimeents: 232. Alcalde-Rico, M., S. Hernando-Amado, P. Blanco and J. L. Martínez (2016). "Multidrug Efflux Pumps at the Crossroad between Antibiotic Resistance and Bacterial Virulence." Frontiers in Microbiology 7(1483). Ambler, R. P. (1980). "The structure of beta-lactamases." Philos Trans R Soc Lond B Biol Sci 289(1036): 321-331. Aminov, R. I. (2009). "The role of antibiotics and antibiotic resistance in nature." Environmental microbiology 11(12): 2970-2988. Anderl, J. N., M. J. Franklin and P. S. Stewart (2000). "Role of antibiotic penetration limitation in Klebsiella pneumoniae biofilm resistance to ampicillin and ciprofloxacin." Antimicrob Agents Chemother 44(7): 1818-1824. Bănăduc, A., D. Bănăduc, A. Burcea, V. Berg, J. Lyche, S. Oancea, H. Olosutean, M. Iasmina, O. Danci, M. Perju and L. Mugurel Gheorghe (2016). The impact of persistent organic pollutants on freshwater ecosystems and human health. Baquero, F., J.-L. Martínez and R. Cantón (2008). "Antibiotics and antibiotic resistance in water environments." Current Opinion in Biotechnology 19(3): 260-265. Barletta, M., A. R. A. Lima and M. F. Costa (2019). "Distribution, sources and consequences of nutrients, persistent organic pollutants, metals and microplastics in South American estuaries." Science of the Total Environment 651: 1199-1218. Bauernfeind, A., J. M. Casellas, M. Goldberg, M. Holley, R. Jungwirth, P. Mangold, T. Rohnisch, S. Schweighart and R. Wilhelm (1992). "A new plasmidic cefotaximase from patients infected with Salmonella typhimurium." Infection 20(3): 158-163. Bauernfeind, A., H. Grimm and S. Schweighart (1990). "A new plasmidic cefotaximase in a clinical isolate of Escherichia coli." Infection 18(5): 294-298. Bauernfeind, A., I. Stemplinger, R. Jungwirth, S. Ernst and J. M. Casellas (1996). "Sequences of beta-lactamase genes encoding CTX-M-1 (MEN-1) and CTX-M-2 and relationship of their amino acid sequences with those of other beta-lactamases." Antimicrob Agents Chemother 40(2): 509-513. Beceiro, A., M. Tomás and G. Bou (2013). "Antimicrobial Resistance and Virulence: a Successful or Deleterious Association in the Bacterial World?" Clinical Microbiology Reviews 26(2): 185-230.
34
Blaak, H., A. H. van Hoek, C. Veenman, A. E. D. van Leeuwen, G. Lynch, W. M. van Overbeek and A. M. de Roda Husman (2014). "Extended spectrum ß-lactamase-and constitutively AmpC-producing Enterobacteriaceae on fresh produce and in the agricultural environment." International journal of food microbiology 168: 8-16. Blair, J. M. A., M. A. Webber, A. J. Baylay, D. O. Ogbolu and L. J. V. Piddock (2014). "Molecular mechanisms of antibiotic resistance." Nature Reviews Microbiology 13: 42. Bush, K. and G. A. Jacoby (2010). "Updated functional classification of beta-lactamases." Antimicrobial Agents and Chemotherapy 54(3): 969-976. Bush, K. K. (2018). "Past and Present Perspectives on beta-Lactamases." Antimicrobial Agents and Chemotherapy 62(10). Cabello, F. C., A. Tomova, L. Ivanova and H. P. Godfrey (2017). "Aquaculture and mcr colistin resistance determinants." MBio 8(5): e01229-01217. Cavallo, J. D., R. Fabre, F. Jehl, C. Rapp and E. Garrabé (2004). "Beta-lactam antibiotics." EMC - Maladies Infectieuses 1(3): 129-202. Chopra, I. and M. Roberts (2001). "Tetracycline Antibiotics: Mode of Action, Applications, Molecular Biology, and Epidemiology of Bacterial Resistance." Microbiology and Molecular Biology Reviews 65(2): 232-260. Courvalin, P. and R. Leclercq (2006). Antibiogramme, Editions Eska. Covich, A., K. Ewel, P. Giller, W. Goedkoop, R. Hall and D. Merritt (2004). Ecosystem services provided by freshwater benthos: 45-73. D'Costa, V. M., C. E. King, L. Kalan, M. Morar, W. W. L. Sung, C. Schwarz, D. Froese, G. Zazula, F. Calmels, R. Debruyne, G. B. Golding, H. N. Poinar and G. D. Wright (2011). "Antibiotic resistance is ancient." Nature 477(7365): 457-461. Davies, J., G. B. Spiegelman and G. Yim (2006). "The world of subinhibitory antibiotic concentrations." Current Opinion in Microbiology 9(5): 445-453. Davies, S. C. (2013). Report of the chief medical officer, Vol 2. . London, Department of Health. de Sá, L. C., M. Oliveira, F. Ribeiro, T. L. Rocha and M. N. Futter (2018). "Studies of the effects of microplastics on aquatic organisms: What do we know and where should we focus our efforts in the future?" Science of The Total Environment 645: 1029-1039. Dudgeon, D., A. H. Arthington, M. O. Gessner, Z. Kawabata, D. J. Knowler, C. Leveque, R. J. Naiman, A. H. Prieur-Richard, D. Soto, M. L. Stiassny and C. A. Sullivan (2006).
35
"Freshwater biodiversity: importance, threats, status and conservation challenges." Biol Rev Camb Philos Soc 81(2): 163-182. Dulio, V., B. van Bavel, E. Brorström-Lundén, J. Harmsen, J. Hollender, M. Schlabach, J. Slobodnik, K. Thomas and J. Koschorreck (2018). "Emerging pollutants in the EU: 10 years of NORMAN in support of environmental policies and regulations." Environmental Sciences Europe 30(1): 5. Fajardo, A., J. F. Linares and J. L. Martínez (2009). "Towards an ecological approach to antibiotics and antibiotic resistance genes." Clinical Microbiology and Infection 15: 14-16. Filutowicz, M., R. Burgess, R. L. Gamelli, J. A. Heinemann, B. Kurenbach, S. A. Rakowski and R. Shankar (2008). "Bacterial conjugation-based antimicrobial agents." Plasmid 60(1): 38-44. Foster, P. L. (2000). "Adaptive mutation: implications for evolution." BioEssays : news and reviews in molecular, cellular and developmental biology 22(12): 1067-1074. Frost, L. S., R. Leplae, A. O. Summers and A. Toussaint (2005). "Mobile genetic elements: the agents of open source evolution." Nature Reviews Microbiology 3(9): 722. Fu, F. L. and Q. Wang (2011). "Removal of heavy metal ions from wastewaters: A review." Journal of Environmental Management 92(3): 407-418. Gall, A. M., B. J. Mariñas, Y. Lu and J. L. Shisler (2015). "Waterborne Viruses: A Barrier to Safe Drinking Water." PLoS pathogens 11(6): e1004867-e1004867. Hall, B. G. and M. Barlow (2004). "Evolution of the serine β-lactamases: past, present and future." Drug Resistance Updates 7(2): 111-123. Heisler, J., P. M. Glibert, J. M. Burkholder, D. M. Anderson, W. Cochlan, W. C. Dennison, Q. Dortch, C. J. Gobler, C. A. Heil, E. Humphries, A. Lewitus, R. Magnien, H. G. Marshall, K. Sellner, D. A. Stockwell, D. K. Stoecker and M. Suddleson (2008). "Eutrophication and harmful algal blooms: A scientific consensus." Harmful Algae 8(1): 3-13. Hyseni, S. (2016) "Toxicological Effects of Nanomaterials on Aqueous and Terrestrial Ecosystems." Center for Development and Strategy 2016. Jacquiod, S., I. Nunes, A. Brejnrod, M. A. Hansen, P. E. Holm, A. Johansen, K. K. Brandt, A. Priemé and S. J. Sørensen (2018). "Long-term soil metal exposure impaired temporal variation in microbial metatranscriptomes and enriched active phages." Microbiome 6(1): 223.
36
Johnson, A. P. and N. Woodford (2013). "Global spread of antibiotic resistance: the example of New Delhi metallo-β-lactamase (NDM)-mediated carbapenem resistance." Journal of medical microbiology 62(4): 499-513. Jones, K. C. and P. de Voogt (1999). "Persistent organic pollutants (POPs): state of the science." Environmental Pollution 100(1): 209-221. Kardol, P., N. Fanin and D. A. Wardle (2018). "Long-term effects of species loss on community properties across contrasting ecosystems." Nature 557(7707): 710-713. Karkman, A., K. Parnanen and D. G. J. Larsson (2019). "Fecal pollution can explain antibiotic resistance gene abundances in anthropogenically impacted environments." Nat Commun 10(1): 80. Kumarasamy, K. K., M. A. Toleman, T. R. Walsh, J. Bagaria, F. Butt, R. Balakrishnan, U. Chaudhary, M. Doumith, C. G. Giske and S. Irfan (2010). "Emergence of a new antibiotic resistance mechanism in India, Pakistan, and the UK: a molecular, biological, and epidemiological study." The Lancet infectious diseases 10(9): 597-602. Laffite, A., D. M. M. Al Salah, V. I. Slaveykova and J. Poté (2019). "Prevalence of β-Lactam and Sulfonamide Resistance Genes in a Freshwater Reservoir, Lake Brêt, Switzerland." Exposure and Health. Lakin, S. M., C. Dean, N. R. Noyes, A. Dettenwanger, A. S. Ross, E. Doster, P. Rovira, Z. Abdo, K. L. Jones, J. Ruiz, K. E. Belk, P. S. Morley and C. Boucher (2017). "MEGARes: an antimicrobial resistance database for high throughput sequencing." Nucleic Acids Res 45(D1): D574-D580. Lazarova, V. and A. Bahri (2005). "Irrigation with recycled water: agriculture, turfgrass and landscape." Lods-Crozet, B., M. De La Harpe, O. Reymond and A. Stawczynski (2009). "Evaluation da la qualité chimique et biologique d’un petit lac du plateau suisse (lac de Bret, canton de Vaud)." Bulletin de la Société vaudoise des Sciences naturelles 91(4): 363-387. Mara, D. and G. Alabaster (2008). "A new paradigm for low-cost urban water supplies and sanitation in developing countries." Water Policy 10(2): 119-129. Margot, J., L. Rossi, D. A. Barry and C. Holliger (2015). "A review of the fate of micropollutants in wastewater treatment plants." Wiley Interdisciplinary Reviews: Water 2(5): 457-487. Marshall, B. M. and S. B. Levy (2011). "Food animals and antimicrobials: impacts on human health." Clin Microbiol Rev 24(4): 718-733.
37
Merlin, C. and A. Toussaint (1999). "Les éléments transposables bacteriens." Société Française de Génétique 15. Minamata Disease Municipal Museum (1994). Ten Things to know about Minamata Disease, The Minamata Environmental Creation Development Namasivayam, C. and K. Kadirvelu (1999). "Uptake of mercury (II) from wastewater by activated carbon from an unwanted agricultural solid by-product: coirpith." Carbon 37(1): 79-84. Nordmann, P., L. Poirel, T. R. Walsh and D. M. Livermore (2011). "The emerging NDM carbapenemases." Trends in Microbiology 19(12): 588-595. NORMAN. (2005). "Network of reference laboratories, research centres and related organisations for monitoring of emerging environmental substances " Retrieved 12/4/2019, 2019, from https://www.norman-network.net/?q=node/19. Nzolo, D. D., P. R. Mulungo and a. l. s. d. p. SIAPS (2013). Évaluation de l’utilisation des médicaments dans six hôpitaux pilotes où les CPT sont fonctionnels en République démocratique du Congo. Oki, T. and S. Kanae (2006). "Global Hydrological Cycles and World Water Resources." Science 313(5790): 1068-1072. Pan, Q., D. Tian, S. Naeem, K. Auerswald, J. J. Elser, Y. Bai, J. Huang, Q. Wang, H. Wang, J. Wu and X. Han (2016). "Effects of functional diversity loss on ecosystem functions are influenced by compensation." Ecology 97(9): 2293-2302. Poff, N. L., J. D. Allan, M. A. Palmer, D. D. Hart, B. D. Richter, A. H. Arthington, K. H. Rogers, J. L. Meyer and J. A. Stanford (2003). "River flows and water wars: emerging science for environmental decision making." Frontiers in Ecology and the Environment 1(6): 298-306. Randall, C. P., K. R. Mariner, I. Chopra and A. J. O'Neill (2013). "The target of daptomycin is absent from Escherichia coli and other gram-negative pathogens." Antimicrobial agents and chemotherapy 57(1): 637-639. Rees, G. and J. Bartram (2002). Monitoring bathing waters: a practical guide to the design and implementation of assessments and monitoring programmes, CRC Press. Roccotiello, E., P. Marescotti, S. Di Piazza, G. Cecchi, M. Mariotti and M. Zotti (2015). Biodiversity in metal-contaminated sites–problem and perspective–a case study. Biodiversity in Ecosystems-Linking Structure and Function, IntechOpen. Schroeder, M., B. D. Brooks and A. E. Brooks (2017). "The Complex Relationship between Virulence and Antibiotic Resistance." Genes 8(1): 39.
38
Singer, A. C., H. Shaw, V. Rhodes and A. Hart (2016). "Review of Antimicrobial Resistance in the Environment and Its Relevance to Environmental Regulators." 7(1728). Srikanth, K. (2019). "Emerging Contaminants Effect on Aquatic Ecosystem: Human Health Risks." Agricultural Research & Technology: Open Access Journal 19(4). Stockholm Convention. "Protecting human health and the environment from persistance organic pollutants." Retrieved 12/04/2019, 2019, from http://www.pops.int/Home/tabid/2121/Default.aspx. Sun, S., M. Selmer and D. I. Andersson (2014). "Resistance to β-lactam antibiotics conferred by point mutations in penicillin-binding proteins PBP3, PBP4 and PBP6 in Salmonella enterica." PloS one 9(5): e97202-e97202. Thevenon, F., L. F. de Alencastro, J.-L. Loizeau, T. Adatte, D. Grandjean, W. Wildi and J. Poté (2013). "A high-resolution historical sediment record of nutrients, trace elements and organochlorines (DDT and PCB) deposition in a drinking water reservoir (Lake Brêt, Switzerland) points at local and regional pollutant sources." Chemosphere 90(9): 2444-2452. Thomas, C. M. and K. M. Nielsen (2005). "Mechanisms of, and barriers to, horizontal gene transfer between bacteria." Nat Rev Microbiol 3(9): 711-721. Tovar-Sánchez, E., I. Hernández-Plata, M. S. Martínez, L. Valencia-Cuevas and P. M. Galante (2018). Heavy Metal Pollution as a Biodiversity Threat. Heavy Metals, IntechOpen. U.S. Geological Survey. (02/12/2016). "The World's Water." Retrieved 10/04/2019, from https://water.usgs.gov/edu/earthwherewater.html. UN-Water. "Monitoring - Sustainable Development Goal #6." Retrieved 11/04/2019, 2019, from http://www.sdg6monitoring.org/. UN Water. (2014). "International Decade for Action "Water for Life" 2005-2015." Retrieved 10/04/2019, 2019, from https://www.un.org/waterforlifedecade/human_right_to_water.shtml. UNEP (2010). Africa Water Atlas. Nairobi, Kenya, Division of Early Warning and Assessment (DEWA). United Nations Environment Programme (UNEP). UNEP/WHO (1996). Water quality monitoring : a practical guide to the design and implementation of freshwater quality studies and monitoring programs. London, E & FN Spon. Vajda, A. M., L. B. Barber, J. L. Gray, E. M. Lopez, J. D. Woodling and D. O. Norris (2008). "Reproductive disruption in fish downstream from an Estrogenic wastewater effluent." Environmental Science & Technology 42(9): 3407-3414.
39
Ventola, C. L. (2015). "The Antibiotic Resistance Crisis: Part 1: Causes and Threats." Pharmacy and Therapeutics 40(4): 277-283. Walsh, F. (2013). "Investigating antibiotic resistance in non-clinical environments." Frontiers in Microbiology 4: 5. WHO (2014). Antimicrobial resistance: global report on surveillance 2014: 257. WHO/EMP (2017). Prioritization of pathogens to guide discovery, research and development of new antibiotics for drug-resistant bacterial infections, including tuberculosis. Geneva. World Meteorological Organization (WMO) (2013). Planning for water-quality monitoring systems, WMO-No. 1113. Geneva, Switzerland, WMO. Wu, W., Y. Feng, G. Tang, F. Qiao, A. McNally and Z. Zong (2019). "NDM Metallo-β- Lactamases and Their Bacterial Producers in Health Care Settings." Clinical Microbiology Reviews 32(2): e00115-00118.
40
CHAPTER 2
Accumulation of toxic metals and organic
micro-pollutants in sediments from tropical
urban rivers, Kinshasa, Democratic Republic of
the Congo
A similar version of this chapter was published under the following reference:
Kilunga, P. I., P. Sivalingam, A. Laffite, D. Grandjean, C. K. Mulaji, L. F. de Alencastro, P. T. Mpiana and J. Poté (2017). "Accumulation of toxic metals and organic micro-pollutants in sediments from tropical urban rivers, Kinshasa, Democratic Republic of the Congo." Chemosphere 179: 37-48. DOI: 10.1016/j.chemosphere.2017.03.081
41
Abstract
The increasing contamination of freshwater resource by toxic metals and Persistent Organic Pollutants (POPs) is a major environmental concern globally. In the present investigation, surface sediments collected from three main rivers named, Makelele, Kalamu and Nsanga, draining through the city of Kinshasa, Democratic Republic of the Congo, were characterized for grain size, organic matter, toxic metals, POPs (including organochlorine pesticides (OCPs), polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs)), and polycyclic aromatic hydrocarbons (PAHs). Furthermore, enrichment factor (EF) and geoaccumulation index (Igeo) were performed to determine metal source and pollution status. The results highlighted high concentration of toxic metals in all sediment samples, reaching the values (mg kg-1) of 325 (Cu), 549 (Zn), 165 (Pb) and 1.5 (Cd) in Makelele. This research provides not only a first baseline information on the extent of contamination in this tropical ecosystem but also represents useful tools incorporated to evaluate sediment quality in the river receiving systems which can be applied to similar aquatic environments.
42
2.1 Introduction
The water resource contamination by toxic metals and POPs is a worldwide problem because these chemicals are not degradable in the environment and can persist in sediments for decades or even centuries. Most of them are characterized by long-term stability and can have high toxic effects on aquatic living organisms (Wildi et al. 2004; Ghrefat and Yusuf, 2006; Thevenon et al. 2012; Mwanamoki et al. 2014b). Previous studies have highlighted that sediment are recipients and reservoirs of toxic heavy metals (Pote et al. 2008; Varol, 2011). In addition, accumulated toxic elements and POPs in sediments over the period of time serve as important indicators to assess and reevaluate the pollution history (Mwanamoki et al. 2014a; Devarajan et al. 2015b; Doong et al. 2008). On the other hand, polluted sediments represent a significant source of contamination in freshwater organisms and have long-term implications for human health (Thevenon et al. 2013; Raghunath et al. 1999). The discharge of untreated urban effluents into river environments is a major concern in developing countries. Given this fact, in recent years accumulation of heavy metals in river sediments from developing countries have been reported with more attention (Mubedi et al. 2013; Devarajan et al. 2015b; Tamim et al. 2016; Laffite et al. 2016). Hydrophobic organic compounds (HOCs), such as PAHs, PCBs, and OCPs have been identified as environmental pollutants in all environmental compartments (Wu et al. 1999). Due to their high persistence and low solubility in water, HOCs can accumulate in sediments (Poté et al. 2008). European Union (EU) and the US Environmental Protection Agency (USEPA) highlighted that PAHs are of significant concern with regard to human health as having carcinogenic properties and bioavailability with water, soil, and sediments (Sindermann, 2006; Zhang et al. 2012). PCBs, PAHs, OCPs, and PBDEs are known to have extraordinary stability, high toxicity, extremely high long-range atmospheric transportability, and potential threats to human health and environmental ecosystems (Cui et al. 2016; Pozo et al. 2012). Heavy metals, POPs and PAHs could be accumulated in aquatic organisms and eventually may transfer to higher order organisms including humans (Pardos et al. 2004; Huang et al. 2006; Díez et al. 2009). Therefore, it is important to assess the accumulation of toxic heavy metals and POPs in the environmental compartments to evaluate the ecological risk.
Kinshasa is the capital and largest city of the Democratic Republic of the Congo (DRC) and has an estimated population of more than 12 million. In Congo DR, urban rivers are specially considered as several sources of pollution including sanitary facilities, landfills, mining activities, discharge of effluents from industries, hospitals, and urban activities. The
43
Makelele, Kalamu, and Nsanga Rivers are the main rivers and tributaries of Congo River that drain the capital city of Kinshasa (Tshibanda et al. 2014; Mwanamoki et al. 2015). These rivers serve as sources of recreational use, bathing, drinking water supply and irrigation for urban agriculture. A very few comprehensive studies of heavy metals, pesticides and POPs in Congo River Basin have been conducted (Verhaert et al. 2013; Mwanamoki et al. 2014b; 2015; Laffite et al., 2016). These studies recommended further researches in the urban river receiving systems in studied area to evaluate the quality of the aquatic ecosystem. The levels of toxic metals, persistent organic pollutants (POPs: including organochlorine pesticides (OCPs), polychlorinated biphenyls (PCBs) and polybrominated biphenyl ethers (PBDEs), and polycyclic aromatic hydrocarbons (PAHs) in sediments are good indicators to evaluate the environmental quality of aquatic systems. Therefore, the objective of the present study was to discuss the occurrence and spatial distribution of toxic metals, POPs and PAHs in sediments from three of main rivers draining the capital city of Kinshasa. Sediment analyses were performed for the physicochemical characterization including sediment grain-size, total organic matter (loss on ignition), total carbon (TC), total phosphorus (TP), metals including Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, Ag, Cd, Sn, Sb, and Pb, and persistent organic pollutants (including OCPs, PCBs, PBDEs) and PAHs. In addition, the degree of sediment pollution by heavy metals was evaluated using geo-accumulation index (Igeo) and enrichment factor (EF) calculation.
2.2 Material and methods 2.2.1 Study sites and sampling procedure Three rivers named Makelele, Kalamu and Nsanga draining the capital city of Kinshasa (Figure 2-1), DRC were selected in this study according to the recommendations of our previous studies (Mubedi et al. 2013; Tshibanda et al. 2014; Mwanamoki et al. 2014ab; 2015). Sampling took place in January 2016. The surface sediments (0-4 cm layer) were collected from (i) Makelele River (R1, n=3, labelled: R1A, R1B, R1C), (ii) Kalamu River (R2, n=4, labelled: R2A, R2B, R2C, R2D) and Nsanga River (R3, n=2, labelled: R3A, R3B). GPS location of sampling site is presented in
Table 2-1.
44
Figure 2-1 Location map of the study area. A: Location Map of Congo DR in Africa. B: Map of Congo DR. C: Location map of studied Rivers, R1: Makelele, R2: Kalamu, R3: Nsanga at Kinshasa, Congo DR.
Approximately 400-500 g of sediments were taken from each site in triplicate. The surface sediments from all sites were collected manually at about 0.3-1 m water depth. Sediment samples were kept in autoclaved glasses for POPs and PAHs analysis and in polyethylene bottles for other analyses. All samples were stored in an icebox at 4 ºC and were transported to the laboratory for different treatments within 24 h. After preliminary treatments, the samples were sent to the Department F.-A. Forel, University of Geneva for analysis.
2.2.2 Sediment grain size and organic matter, total organic carbon, total nitrogen and phosphorus analysis The particle grain size was measured in fresh sediment with a Laser Coulter® LS-100 diffractometer (Beckman Coulter, Fullerton, CA, USA), following 5 min ultrasonic dispersal in deionized water. The sediment total organic matter content was estimated from mass loss on ignition, at 550°C during 1 hour in a Salvis oven (Salvis AG, Luzern, Switzerland).
45
The percentage of TC and Total Nitrogen (TN) was measured with an Elemental Analyzer (Perkin Elmer CHNS/O PE 2400 Series II, Waltham, MA, USA) by the following conditions: Combustion Temperature: 975°C, Reduction Temperature: 500°C, Detection Oven: 82.2 °C, Pressure: 514.0 mm Hg.
The TP was measured with a spectrophotometer (Helios Gamma UV-Vis, Thermo scientific, Waltham, USA) at 850 nm. Fifty milligrams of dry sediments were diluted in 5 mL 1N HCl in centrifuge tubes and ultrasonicated (at ambient temperature) for 16 h and then centrifuged at 4000 rpm for 20 min. The supernatant was mineralized for 45 min at 130°C after addition of 5 % K2S2O8 solution. The TP concentration was determined by measuring the absorbance of the blue complex obtained after reduction of molybdophosphoric acid according to the method described by Poté et al. (2008).
2.2.3 Metal analysis in sediment samples Before analysis, the sediment samples were lyophilized, grounded into a fine powder, homogenized, sieved through a 63 µm mesh size sieve and digested according to the previous method as described by Poté et al. (2008). The digested samples were subjected to analysis by Inductive Coupled Plasma-Mass Spectroscopy (Agilent 7700x series ICP-MS developed for complex matrix analysis, Santa Clara, CA, USA). A collision/reaction cell (Helium mode) and interference equations were utilized to remove spectral interferences that might otherwise bias results. This is sufficient for many routine applications. Multi-element standard solutions at different concentrations (0, 0.02, 1, 5, 20, 100 and 200 µg L-1) were used for calibration. The certified sediment reference materials LKSD-2 and LKSD-4 were used for lake and river sediment analysis in order to verify the sensibility of the instrument and the reliability of the results, respectively. Concentrations are in mg kg-1 (ppm) dry weight. The standard deviations of 3 replicate measurements were below 5 %, and chemical blanks for the procedure were less than 1 % of the sample signal.
2.2.4 Geoaccumulation index and enrichment factor The enrichment factor (EF) and geoaccumulation index (Igeo) in sediment samples were calculated as described by Maanan et al. (2004), Varol (2011) and Thevenon et al. (2012, 2013). The Igeo accumulation index is defined by the following equation:
46
퐶푛 퐼푔푒표 = 푙표푔 2 ( ) 1.5 퐵푛
n: concentration of metals (n) examined in the sediment samples
Bn: concentration of the metal (n) geochemical background
1.5: lithospheric effect background correlation matrix factor
Enrichment factor is a useful tool to determine the degree of anthropogenic heavy metal pollution. EF is calculated using the following equation, and according to our previous study, Scandium (Sc) was used for the geochemical normalization (Mwanamoki 2014b).
푚푒푡푎푙 ( 푆푐 ) 퐸퐹 = 푠푎푚푝푙푒 푚푒푡푎푙 ( ) 푆푐 푏푎푐푘푔푟표푢푛푑
2.2.5 Chlorinated pesticides, PCBs, PAHs and PBDEs analysis Chlorinated pesticides, PCBs, PAHs and PBDEs analysis was performed according to our previous paper as described by Mwanamoki et al. (2014b) and Thevenon et al. (2013). Briefly: all glassware was rinsed with acetone and hexane and tested before using. Blanks do not present any quantifiable amount of contaminants or interfering compounds. After addition of surrogate standards (13C-labeled for all halogenated compounds and deuterated-labeled compounds for all PAHs), about 5 g of dry sediment were extracted with a mixture of 20 % of acetone in 80 % of hexane (v/v) for 4 hours into a Soxhlet system (Buchi B-811, Flawil, Switzerland). Interfering sulfur compounds were removed by addition of activated copper to the extract. Then, the organic extract was concentrated to 1 mL in a vacuum rotary evaporator (Buchi Rotavapor, Flawil, Switzerland).
The extract was further submitted to fractionation and clean-up over a chromatographic column containing 3 g of Silicagel, according to de Boer et al. (2001). Three separated fractions were collected: first with 16 mL of hexane, then 35 mL of hexane, and finally 50 mL of hexane: dichloromethane (v/v, 1:1). The first two fractions should contain respectively PCBs and PBDEs. PAHs and chlorinated pesticides are distributed on the 3 fractions. After a new reduction of the volume, chemicals were measured by gas chromatography with triple mass
47 spectrometry detection (GC-MS/MS, Thermo Scientific, TSQ Quantum XLS Ultra, Waltham, MA, USA). Two columns with different polarities, a ZB-5ms column (60m x 0.25mm x 0.25µm) and one ZB-XLB column (20m x 0.18mm x 0.18µm), were used for separation and identification of the different compounds. The choice of the column for the identification of a specific compounds is mentioned on Table 2-4 and Table 2-5.
Chlorinated pesticides analyzed in sediment were: hexachlorocyclohexane (HCH) (isomers alpha, beta & gamma) heptachlor, heptachlorepoxyde cis, heptachlorepoxyde trans, hexachlorobenzene , aldrin, dieldrin, endrin, oxy-chlordane, alpha-chlordane, gamma- chlordane, cis-nanochlor, trans-nonachlor, o,p’-DDE (dichloro diphenyl dichlorethylene), p,p’- DDE (dichloro diphenyl dichlorethylene), o,p’-DDD (dichloro diphenyl dichlorethane), p,p’- DDD (dichloro diphenyl dichlorethane), o,p’-DDT (dichloro diphenyl trichlorethane) , p,p’- DDT (dichloro diphenyl trichlorethane) and mirex. In the present work, results for DDT compounds are given as ΣDDT, meaning the sum of the DDT and metabolites. These pesticides are included in the Second Round of UNEP-coordinated Global Interlaboratory Assessment 2012/2013 (UNEP-POPs, 2012). Selected congeners of PCBs searched were IUPAC numbers: CB-28 (2,4,4’-trichlorobiphenyl), CB-52 (2,5,2’,5’- tetrachlorobiphenyl), CB-101 (2,4,5,2’,5’- pentachlorobiphenyl), CB-149 (2,3,6,2’,4’5’-hexachlorobiphenyl), CB-118 (2,4,5,3’,4’- pentachlorobiphenyl), CB-153 (2,4,5,2’,4’,5’-hexachlorobiphenyl), CB-105 (2,3,4,3’,4’- pentachlorobiphenyl), CB-138 (2,3,4,2’,4’,5’- hexachlorobiphenyl), CB-128 (2,3,4,2’,3’,4’- hexachlorobiphenyl), CB-156 (2,3,4,5,3’,4’- hexachlorobiphenyl), CB-180 (2,3,4,5,2’,4’5’- heptachlorobiphenyl) and CB-170 (2,3,4,5,2’,3’,4’- heptachlorobiphenyl). These CBs were used for certification purposes by the former Community Bureau of Reference (BCR) of the European Union (Wells et al. 1992). Results are given as Σ12PCBs meaning the sum of all CBs measured in a sample and also as “Total PCBs” meaning the sum of 7 selected CBs (28, 52, 101, 118, 138, 153 and 180) multiplied by a correction factor of 4.3 (FOEN, 1998).
PBDEs: Selected congeners were IUPAC numbers: BDE-17 (2,2’,4-tribromodiphenyl ether), BDE-28 (2,4,4’-tribromodiphenyl ether), BDE-47 (2,2’,4,4’-tetrabromodiphenyl ether), BDE-66 (2,3’,4,4’-tetrabromodiphenyl ether), BDE-85 (2,2’,3,4,4’-pentabromodiphenyl ether), BDE-99 (2,2’,4,4’,5-pentabromodiphenyl ether), BDE-100 (2,2’,4,4’,6- pentabromodiphenyl ether), BDE-138 (2,2’,3,4,4’,5’-hexabromodiphenyl ether), BDE-153 (2,2’,4,4’,5,5’-hexabromodiphenyl ether), BDE-154 (2,2’,4,4’,5,6’-hexabromodiphenyl ether), BDE-183 (2,2’,3,4,4’,5’,6-heptabromodiphenyl ether), BDE-190 (2,3,3’,4,4’,5,6- heptabromodiphenyl ether). BDE-209 (decabrominated) was not analyzed. ΣPBDEs is the sum
48 of all studied PBDEs. These PBDEs are included in the Second Round of UNEP-coordinated Global Interlaboratory Assessment 2012/2013 (UNEP-POPs, 2012).
Quantitation limits (LOQ) defined as 10 times baseline noise, were comprised between 0.02 ng g-1 (p,p’-DDE) and 0.15 ng g-1 for BDE-190 dry weight.
PAHs: naphthalene, acenaphthene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene, benzo(a)anthracene, chrysene, benzo(e)pyrene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, dibenz(a,h)anthracene, benzo(g,h,i)perylene, indeno (1,2,3c,d)pyrene, ΣPAHs means the sum of all 16 studied PAHs. These PAHs are similar to those proposed by US EPA to be measured in sediment samples, with exception of benzo(e)pyrene and addition of acenaphtylene.
The results for POPs and PAHs analysis are expressed in µg kg-1 dry weight.
2.2.6 Data analysis Triplicate measurements were performed for all the analyses. Statistical treatment of data (Spearman’s rank order correlation) has been realized using SigmaStat 12.5 (Systat Software, Inc., USA). Principal Component Analysis (PCA), a multivariate statistical analysis was performed using R (R Core Team, 2015) in order to understand relationship among analyzed compound and their potential sources. Prior to performing PCA analysis, data were centered in order to maximize the dispersion.
2.3 Results and discussion 2.3.1 Physicochemical characteristics of sediments Sediment characteristics including particle grain size, OM, TC, TN, and TP are shown in Table 2-1. The total OM in sediments ranged from 5.3 to 9.2, 7.2-12.6 and 1.7-13.4% for the sites R1, R2, and R3, respectively. There was no significant difference in total OM among all samples (triplicate) from the same site. According to the results of our previous studies (Poté et al. 2008; Haller et al. 2009; Mubedi et al. 2013), the organic matter in non-contaminated freshwater sediments varied from 0.1-6.0%. The results of this study indicated that the sediment from studied sites can be considered as polluted/moderately polluted by organic matter. For
49
example, OM can reach more than 30% in sediments contaminated by the municipal WWTP effluent waters (Poté et al. 2008; Devarajan et al. 2015a).
Table 2-1 GPS location of sampling sites and physico-chemical parameters of surface sediments from Makelele, Kalamu and Nsanga Rivers
Median Site Sample Longitude Latitude OM grain size Clay Silt Sand TC TN TP (%) (µm) (%) (%) (%) (%) (mg kg-1 ) (mg kg-1)
R1A 15°16’28.0’’ 4°19’95.0’’ 9.15 105.3 2.06 29.7 68.2 1.65 2.56 141.8 R1 R1B 15°16'24.5’’ 4°19’89.0’’ 5.26 101.8 2.08 21.2 76.6 2.80 3.04 92.6 R1C 15°16’29.2’’ 4°19'94.7’’ 8.05 46.7 4.08 48.0 47.9 4.79 6.32 242.7 R2A 15°19’38.0’’ 4°21’08.0’’ 7.19 191.6 0.60 12.7 86.6 2.10 3.08 98.6 R2 R2B 15°19’37.0’’ 4°21’11.2’’ 12.64 32.8 2.64 68.1 29.2 3.62 2.80 76.4 R2C 15°19’37.3’’ 4°21’08.1’’ 7.95 195.9 0.27 9.0 90.6 7.20 3.20 133.6 R2D 15°19’37.1’’ 4°21’06.8’’ 11.38 50.8 0.74 52.5 46.7 2.50 3.60 163.2 R3A 15°22’49.6’’ 4°23’23.4’’ 1.73 225.3 1.50 6.6 91.8 1.72 4.20 89.7 R3 R3B 15°22’49.2’’ 4°23’23.9’’ 13.35 79.5 0.80 40.7 58.4 2.37 1.78 67.4 R1, R2, R3: Makelele, Kalamu, Nsanga Rivers, respectively. MO: total organic matter, TC: total carbon, TN: total nitrogen, TP: total phosphorus
Surface sediments of rivers in all studied sites are generally sandy-silt. For all samples, the sand values ranged from 61.0-95.0% and silt from 5.8-39.0%. The maximum value of clay was observed at the site R2A (1.5%). The sediment median grain size varied substantially internally within the sampling sites (p˂0.05). The values ranged from 46.8-105.3, 32.9-195.9 and 79.5-225.3 μm for R1, R2 and R3 respectively. In accordance with our previous report, the percentage of grain size in sediments observed in this study could probably be responsible for high porosity and high permeability, and further flexible transportation of the sediment downstream (Wildi et al. 2004; Thevenon et al. 2011; Bartoli et al. 2012). The values of TC content varied from 1.7-4.8, 2.1-7.2 and 1.7-2.4 % for R1, R2, and R3, respectively. The TN exhibited relatively low concentrations and a limited variability in sediments. The values ranged from 2.6-6.3, 2.8-3.6 and 1.8-4.2 mg kg-1 for the sites R1, R2, and R3, respectively. The total phosphorous contents of the sediments varied considerably with sampling sites (p˂0.05). The values ranged from 92.6-242.7, 76.5-163.2, 67.4-89.8 mg kg-1 for the sites R1, R2, and R3, respectively. The results obtained are in agreement with our previous studies, which demonstrated that there are large variations in the distribution of TC, TN, TP and grain size in sediments from some urban rivers located in the city of Kinshasa (Mubedi et al. 2013; Tshibanda et al. 2014; Kilunga et al. 2016; Laffite et al. 2016).
50
2.3.2 Metal concentrations in the surface sediments The results of the heavy metals analysis are presented in Table 2-2. It is noted that the concentrations of four heavy metals including Cu, Zn, Cd and Pb were found considerably higher in most of the sediment samples. The concentration of Zn was found to be higher than the threshold level in all sediment samples and the value ranged from 128.1-549.6 mg kg-1. The concentration of Pb was found to be in the range of 24.6-165.3 mg kg-1. It was observed that the concentration of Pb generally high in all sediments except for the site R2A. The highest concentration of Cu was recorded at the site R3B with the value of 325.14 mg kg-1. Similarly, for Cd at 1.5 mg kg-1 and Pb at 165.30 mg kg-1 in the same site. The concentration of heavy metals including Cu, Zn, Cd and Pb recorded in this study were primarily compared with the Sediment Quality Guidelines for the Protection of Aquatic Life (CCME EPC-98E 1999). The evaluation of the potentially deleterious effects of the metals towards benthic fauna, which is based on consensus-based guidelines for the sediment quality (MacDonald et al. 2000a; Long et al. 2006), can give an estimate of the hazard that the sediments may represent for the local biota. Authors proposed (MacDonald et al. 2000a; Long et al. 2006) for specific metals a ‘‘threshold effect concentration’’ (TEC), a level above which some effect (or response) will be produced in an organism and below which it will not, and a ‘‘probable effect concentration’’ (PEC), a contaminant level that is likely to cause an adverse effect on biota. Though distribution of heavy metals varied among all sampling sites, the concentration of Cu, Zn, Cd, and Pb is generally higher than the threshold level. Surprisingly, the concentrations of Zn, Pb and Cd are higher than that of the Congo River (Mwanamoki et al. 2014b; Mwanamoki et al. 2015). The results of the present study show that the higher concentrations of Cu, Ni, Zn, and Pb are likely to have a harmful effect on aquatic organisms. Therefore, the sediments from Makelele, Kalamu and Nsanga Rivers can be considered as highly polluted by heavy metals and this could be explained by various industrial sources and urban discharge into the rivers. Urban rivers in Kinshasa seem to be under high threat and stretch of the river running through urban areas is extremely polluted due to open drains, sewage inflow and landfills (Mavakala et al. 2016; Mwanamoki et al. 2015, Mubedi et al. 2013). However, the presence of other non-identified sources (such as artisanal activities) and untreated hospital effluent water discharge cannot be excluded (Laffite et al. 2016).
51
Table 2-2 Metal content of surface sediment samples from River Makelele (R1), Kalamu (R2) and Nsanga (R3) analyzed by ICP-MSa
Concentration (mg.kg-1)
Sampling Sample sites number Sc Ti Cr Co Ni Cu Zn As Mo Ag Cd Sn Sb Pb Makelele R1A 0.71 63.93 12.57 0.96 3.94 31.68 177.15 1.03 0.32 0.10 0.76 1.89 0.20 38.03 R1B 0.37 41.67 8.43 0.71 2.01 18.19 128.10 0.49 0.08 0.07 0.23 0.49 0.22 37.80 R1C 1.30 98.89 28.51 3.93 9.80 71.91 549.64 1.64 0.22 0.31 1.26 1.61 0.32 132.29 Kalamu R2A 0.22 49.67 6.67 1.27 2.95 23.52 304.68 0.83 0.00 0.69 1.13 0.97 0.60 64.49 R2B 0.26 54.31 10.14 1.30 4.04 39.79 333.54 0.65 0.06 0.21 0.86 0.60 0.77 71.20 R2C 0.12 30.55 4.77 0.68 2.27 33.04 177.62 0.44 n.a 0.07 0.41 0.57 0.69 65.83 R2A 0.21 50.67 8.45 1.24 4.33 37.55 286.76 0.55 0.13 0.23 0.78 0.88 0.66 64.57 Nsanga R3A 0.12 26.36 7.51 0.53 11.18 213.03 130.17 0.29 0.17 0.32 0.48 1.47 0.49 24.59 R3B 0.18 53.38 14.01 1.41 16.94 325.14 396.72 0.88 0.54 0.54 1.54 1.98 1.47 165.30 Recb.max 37.30 35.70 123.00 5.90 0.60 35.00 LKSD 4 21 11 32 30 189 2 0.2 1.9 1.5 93 a Total variation coefficients for triplicate measurements are smaller than 5 % for ICP-MS analysis. The recovery values from the ICP-MS triplicate measurements for reference material (LKSD 4) was above 97.5 % for all elements. b Canadian Sediment Quality Guidelines for the Protection of Aquatic Life recommendation. In bold represent the concentration of the heavy metals above the recommended concentration according to the Canadian Sediment Quality Guidelines for the Protection of Aquatic Life recommendation (CCME EPC-98E 1999). n.a – analysis not performed.
2.3.3 Enrichment factor (EF) and Geoaccumulation index (Igeo) EF and Igeo values for selected metals in sediment samples from rivers are presented in Table 2-3. The EF and Igeo indices are important in discriminating between anthropogenic metals and provide a quantitative criterion for characterizing the sediment according to the degree of metal pollution (Adamo et al. 2005). According to the Igeo values, pollution of toxic metals in studied sites was classified in the order of Zn > Pb > and Cu. The Igeo classification of “extremely polluted” was observed for Zn in samples from the sites R2C and R3B, and “heavily polluted” were observed in samples from R1A, R1B, and R2A. Other sites are “moderately to heavily polluted” level for Zn. For Pb, “moderately to heavily polluted” level was observed for all samples for the river R1, while the site R2C and R3B present “heavily polluted” level. Cu, Igeo values were ranged from -0.40 to 2.63 and the high Igeo value was observed in the site R3A. Based on the Igeo values of Cu, except for the sites R2B and R2C, other sites were not polluted by Cu.
52
Table 2-3 Igeo and EF values for Cu, Zn and Pb in surface sediments.
Igeo EF Sampling sites Sample number Cu Zn Pb Cu Zn Pb R1A -1.16 3.67 2.62 3.12 88.43 42.78 R1B -0.40 3.80 2.76 4.35 79.69 38.88 Makelele R1C -0.67 2.89 2.65 7.87 92.49 78.35 R2A -0.48 3.58 2.62 5.08 84.80 43.64 Kalamu R2B 2.02 2.44 1.23 51.15 68.38 29.52 R2C 2.63 4.05 3.98 50.68 135.26 128.81 R2D -0.73 2.88 1.86 1.28 15.70 7.70 R3A -1.53 2.42 1.85 1.40 21.62 14.58 Nsanga R3B 0.45 4.52 3.66 1.58 26.44 14.54 Igeo ≤ 0 Class 0 - practically unpolluted EF < 1 no enrichment 0 < Igeo < 1 Class 1 - unpolluted to moderately polluted EF < 3 minor enrichment 1 < Igeo < 2 Class 2 - moderately polluted EF 3 - 5 moderate enrichment 2 < Igeo < 3 Class 3 - moderately to heavily polluted EF 5 - 10 moderately severe enrichment 3 < Igeo < 4 Class 4 - heavily polluted EF 10 - 25 severe enrichment 4 < Igeo < 5 Class 5 - heavily to extremely polluted EF 25 - 50 very severe enrichment 5 > Igeo Class 6 - extremely polluted EF > 50 extremely severe enrichment
The enrichment results were interpreted according to previous studies (e.g. Sakan et al. 2009; Mavakala et al. 2016). The EF values for Cu was ranged from 1.28 to 51.15, indicating “minor to extreme enrichment”. For Cu the highest EF value was recorded at site R2B and R2C, indicating “extreme enrichment”. Whereas, three sites (R2D, R3A, and R3B) for Cu showed lesser values of EF (1.28 to 1.58), implied that there were minor enrichments of Cu. The EF values for Zn ranged from 15.70 to 135.26, indicating “severe to extremely severe enrichment”. Intriguingly, Zn showed the highest EF values among the toxic metals investigated. The EF values of Zn for the sites R1A (88.43), R1B (79.69), R1C (92.49), R2A (84.80), R2B (68.38) and R2C (135.26) indicated “extreme enrichment”. It was observed that the EF values of Pb ranged from 7.70 to 128.81, which indicated that “moderately severe to extremely severe enrichment”. Furthermore, the EF values of Pb revealed “extreme enrichment” for the sites R2C (78.35) and R3A (128.81) and very “severe enrichment” for the sites R1A (42.78), R1B (38.88), R2A (43.64) and R2B (29.52). The results suggested that Zn and Pb could be originated from anthropogenic sources. Whereas Cu might have originated both from natural and anthropogenic sources (Mavakala et al. 2006; Mwanamoki et al. 2015).
53
2.3.4 Spatial distribution of persistent organic pollutants in sediments The concentration of persistent organic pollutants (POPs) including organochlorine pesticides (OCPs), polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and polycyclic aromatic hydrocarbons (PAHs) in surface sediments from river Makelele (R1), Kalamu (R2) and Nsanga (R3) are shown in Table 2-4 and Table 2-5. All results are reported to dry weight. PCBs were detected in most surface sediment samples. In the current study 12 PCB congeners (PCBs 28, 52, 101, 105, 118, 128, 138, 149, 153, 156, 170 and 180) were detected. The sum of 12 PCBs (Σ12 PCBs) ranged between 3.46 to 52.94 µg kg-1. Comparison of the (Σ7 PCBs x 4.3) PCBs for surface sediment samples at the nine sampling sites in three different rivers revealed the maximum concentration for site R1C (169.29 µg kg- 1), followed by R2D (83.24 µg kg-1) and R3B (35.90 µg kg-1). Concentrations increase from upper stream to downstream in all rivers, confirming the enrichment of PCBs coming from more heavily populated areas. In all sediments, the most abundant congeners were in order 153>138>180>149>101, penta and hepta chlorobiphenyls. These congeners are reflecting historic use of PCB mixtures containing penta, hexa and hepta chlorobiphenyls. This pattern is somewhat different from Verhaert et al. (2013) who found a PCB pattern indicating dominant use of arochlor 1254. PCB 138, 153 and 180 are among the most persistent PCB congeners, and have long half-life. As the use of commercial PCBs was forbidden in the 1980s, what we find here are a number of the most persistent PCB congeners, in this case in aquatic sediment environment. Fair enough, the pattern of PCB is reflecting historic use of commercial mixtures but it is also important to stress that the pattern found in this study reflects high persistence (long half-life) (Stockholm Convention, 2017). MacDonald et al. (2000b) proposed values for TEL (59.8 µg kg-1) and PEC (676 µg kg-1) for “total PCBs”, that we can compare with our Σ7 PCBs x 4.3. Site R1C (Makelele River) and all sites (R2A, R2B, R2C and R2D) from Kalamu River presented higher concentrations than PEL value. To understand the degree of contamination, the concentrations of PCBs in sediments from Rivers Makelele, Kalamu and Nsanga were compared with PCB levels recorded in previous studies in Congo DR and on other tropical developing countries.
54
Table 2-4 Concentration (in µg kg-1 dry weight) of polychlorinated biphenyl (PCBs), and polycyclic aromatic hydrocarbons (PAHs) in sediment samples from River Makelele (R1), Kalamu (R2) and Nsanga (R3).
Makelele Kalamu Nsanga PCBs (µg kg-1) LOQ R1A R1B R1C R2A R2B R2C R2D R3A R3B 28a 0.05 0.65 0.54 2.64 0.74 1.50 0.34 1.50 <0.05 0.69 52 0.05 0.73 0.51 2.85 0.56 1.15 0.29 1.21 <0.05 0.97 101 0.10 0.84 0.77 4.48 0.80 1.40 0.73 1.75 <0.10 0.97 105 0.05 0.22 0.27 2.02 0.36 0.48 0.22 0.52 <0.05 0.40 118 0.10 0.63 0.65 4.27 0.58 1.25 0.57 1.56 <0.10 1.11 128 0.05 0.15 0.22 1.68 0.32 0.43 0.33 0.67 <0.05 0.24 138a 0.05 0.76 1.27 7.91 1.65 2.84 2.32 3.83 0.42 1.47 149 0.10 0.63 0.80 5.85 1.44 2.01 2.15 3.18 0.36 1.15 153 0.05 1.17 1.46 10.89 1.94 3.79 3.77 5.39 0.70 2.09 156 0.05 <0.05 <0.05 0.92 0.15 0.32 0.17 0.41 <0.05 0.06 170 0.05 0.23 0.38 3.10 0.86 1.40 1.49 2.02 0.53 0.53 180 0.05 0.37 0.83 6.33 1.68 2.95 3.25 4.12 1.45 1.05 Total PCBs (Σ7 x 4.3) 22.14 25.92 169.29 34.18 63.98 48.46 83.24 11.05 35.90 TELb (Σ7 PCBs) 34.1 PELb (Σ7 PCBs) 277 PAHs (µg kg-1) Naphthalène 0.5 14.61 14.82 103.60 35.90 46.52 98.01 93.95 <0.50 50.45 Acenaphthylene 0.5 1.43 1.65 7.94 3.23 4.03 3.16 5.07 1.15 6.58 Acénaphthène 0.5 1.50 1.20 9.03 3.17 3.44 2.19 3.91 <0.50 14.45 Fluorène 0.5 6.00 5.22 46.41 10.94 20.04 6.67 17.55 <0.50 43.61 Phénanthrène 0.5 27.90 20.72 170.41 45.10 72.95 47.68 84.15 6.51 159.82 Anthracène 0.5 2.82 2.49 25.10 5.66 8.88 5.54 10.45 <0.50 16.85 Fluoranthène 1 20.33 12.83 108.55 29.75 51.11 31.04 48.17 3.56 105.45 Pyrène 1 24.11 19.64 168.99 34.33 59.11 33.69 51.22 4.97 154.30 Benzo (a) Anthracène 1 6.67 5.37 42.40 11.81 19.99 12.18 18.30 <1.00 46.64 Chrysène 1 21.07 10.21 78.71 22.32 35.29 21.50 37.19 2.55 69.87 Benzo(b)Fluoranthène 1 11.75 9.68 71.17 21.91 32.29 21.03 34.47 1.72 82.27 Benzo (k) Fluoranthène 1 3.44 3.42 24.13 6.92 10.27 7.30 12.25 <1.00 30.28 Benzo (a) Pyrène 1 5.86 5.70 39.87 12.66 18.22 11.47 20.07 <1.00 50.03 Dibenz (a,h) Anthracènea* 1 <1.00 <1.00 9.42 3.15 4.22 3.32 4.66 <1.00 12.58 Benzo (g,h,i) Perylène* 1 6.85 9.54 62.67 17.95 25.96 13.97 31.18 2.10 58.45 Indeno (1,2,3c,d) pyrenea* 1 4.74 6.22 43.54 13.47 19.90 11.00 22.69 <1.00 49.50 Σ16 PAH congeners 159.08 128.71 1011.94 278.27 432.22 329.75 495.28 22.56 951.13 TELc (Σ13 PAHs) 610 PELc (Σ13 PAHs) 22800 aChromatographic separation on a ZB-XLB column. All remaining compounds on a ZB-5ms column. bThreshold effect levels (TELs)/Probable effect levels (PELs) (dry weights). Canadian sediment quality guidelines for the protection of aquatic life. Canadian Council of Ministers of the Environment. (CCME, 2002). cMacDonald et al 2000b. Development and evaluation of consensus-based sediment quality guidelines for freshwater eco- systems. *Not included in the sum of (Σ13 PAHs) TEL and PEL values
55
Table 2-5 Concentration (in µg kg-1 dry weight) of organochlorine pesticides (OCPs) and BDEs in sediment samples from river Makelele , Kalamu and Nsanga
Makelele Kalamu Nsanga Reca OCPs (µg kg-1) LOQ R1A R1B R1C R2A R2B R2C R2D R3A R3B hexachlorobenzène 0.05 <0.05 0.40 1.44 <0.05 0.09 <0.05 0.23 0.10 <0.05 alpha-HCH 0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 beta-HCH 0.05 0.10 <0.05 0.14 <0.05 <0.05 0.12 <0.05 <0.05 <0.05 gamma-HCH 0.10 0.26 0.17 0.44 <0.10 0.12 0.43 <0.10 <0.10 <0.10 delta-HCH 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 chlorpyrifos-methylc 1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 chlorpyrifos-ethylc 0.50 1.95 2.30 16.98 1.86 3.68 1.55 1.82 <0.50 3.36 gamma-chlordane 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 alpha-chlordane 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 dieldrin 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 endrin 0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 heptachlor 0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 aldrinc 0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 heptachlor epoxid A 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 heptachlor epoxid B 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 endosulfan I 1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 endosulfan II 1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 endosulfan sulfate 0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 endrin aldehyde 0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 endrin ketone 0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 methoxychlor 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 acetochlorc 1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 <1.00 oxy-chlordanec 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 trans-nonachlor 0.10 <0.10 <0.10 0.20 <0.10 <0.10 <0.10 <0.10 <0.10 <0.10 mirex 0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 <0.30 cyhalothrin-λ (lambda) 1.00 2.67 1.71 12.42 2.29 3.33 <1.00 <1.00 <1.00 <1.00 cypermethrin a 5.00 13.84 <5.00 31.12 10.85 8.17 <5.00 <5.00 <5.00 34.95 cypermethrin b 7.00 16.61 <7.00 20.80 <7.00 <7.00 <7.00 <7.00 <7.00 <7.00 deltamethrin 7.00 43.83 27.31 227.48 42.68 97.33 43.83 32.17 12.49 69.14 DDTs (µg kg-1) p,p'-DDE 0.10 146.75 9.57 63.71 9.54 15.92 11.03 20.99 0.66 9.97 o,p'-DDE 0.10 7.51 0.26 1.72 0.23 0.40 0.33 0.46 <0.10 0.18 p,p'-DDD 0.10 21.57 5.10 39.40 5.84 10.15 4.95 12.79 0.21 4.86 o,p'-DDDc 0.10 22.03 1.47 9.18 1.56 2.59 1.81 3.62 <0.10 1.33 p,p'-DDT 0.10 29.17 4.79 17.14 6.49 2.70 3.73 3.41 0.27 3.09 o,p'-DDTc 0.10 43.58 0.56 1.47 1.05 0.55 0.78 0.81 0.09 0.82 Σ6 pesticides 270.61 21.75 132.62 24.71 32.31 22.63 42.08 1.23 20.25 TELb(Σ6DDTs) 6.15 PELb (Σ6DDTs) 20.03 PBDE (µg kg-1) BDE28 0.15 <0.15 <0.15 <0.15 <0.15 <0.15 <0.15 <0.15 <0.15 <0.15 44 BDE47 0.10 1.80 0.55 4.24 1.21 1.93 0.83 2.27 1.32 8.66 39 BDE100 0.10 0.54 0.18 1.17 0.40 0.54 0.28 0.75 0.51 2.95 0.4 BDE99 0.10 2.23 0.75 5.87 1.27 2.16 0.97 2.55 1.79 12.88 0.4 BDE154 0.15 0.29 <0.15 0.69 <0.15 0.42 <0.15 0.36 0.28 1.48 440 BDE153 0.15 <0.15 <0.15 0.92 <0.15 0.50 <0.15 0.55 <0.15 1.98 440 Σ6 PBDEs 4.86 1.48 12.90 2.88 5.55 2.09 6.48 3.90 27.95
56
The values in bold represent the concentration of the PBDEs above the recommended concentration according to the Canadian Federal Environmental quality guidelines Polybrominated Diphenyl Ethers (PBDEs). (Canadian Environmental Protection Act, 1999. Reca. Federal Environmental quality guidelines Polybrominated Diphenyl Ethers (PBDEs) in sediments.
The contents of (Σ7 PCBs x 4.3) PCBs were higher than those reported for the Congo river basins (Verhaert et al. 2013; Mwanamoki et al. 2014b) and a similar site, Yamuna River in Delhi in India (Kumar et al. 2013). However, concentrations of PCBs were lower as compared to reported by Malik et al. (2014) from the sediment of Soan River, Pakistan (27.9- 116 µg kg-1). In order to investigate potential PCBs sources, PCA was performed (Figure 2-2a). The active variable was composed of the congeners composition of PCBs at different sampling sites. The first component was mainly influenced (94.5%) by the PCBs load. The second component explains the relative congener composition in samples. Based on the loading plot of the PCA, strong differences in PCBs load and relative congener pattern was observed among the different R1, R2, and R3 sampling sites. Furthermore, all the PCBs were correlated with each other (R> 0.77, p<0.05) but three cluster of highly strong correlation between PCB congeners can be seen: 101-105-118 (0.99 57 Figure 2-2 Score plot for principal component analysis (PCA) applied to sediment measurement across sampling sites: (a) PCA of PCBs congeners, (b) PCA of pesticides congeners, (c) PCA of PBDEs congeners and (d) PCA of PAHs congeners For the 35 OCPs targeted in this study, only 14 were detected i.e. hexachlorobenzene, β- and γ-HCH, chlorpyrifos-ethyl, trans-nonachlor, p,p’-and o,p’-DDE, p,p’- and o,p’-DDD, p,p’- and o,p’-DDT, cyhalothrin-λ, cypermethrin a, cypermethrin b and deltamethrin. None of the OCPs α- and δ-HCH, chlorpyrifos-methyl, α-, γ- and oxy-chlordane, dieldrin, endrin, endrin aldehyde and ketone, heptachlor, heptachlor epoxid A and B, aldrin, endosulfan I and II, endosulfan sulfate, methoxychlor, acetochlor and Mirex were detected. The non-detection of these OCPs may be due to their regulation and/or banishment by US EPA and European directives. Figure 2-2b shows the results of PCA analysis for OCPs. The first and second components explained 59.7% and 27.5% of the total variance. The first component was mainly influenced by the quantity of OCPs while the second component was influenced by congeners composition. All sites are grouped in the same cluster in reason of a similar OCPs mixture, but 58 R1A and R1C differ from the other sites because of their strong contamination by DDT and its derivatives at R1A site and chlorpyrifos-ethyl & deltamethrin at the R1C site. Pyrethroids pesticides like cypermethrin, cyhalothrin and deltamethrin have all neurotoxic effects and may cause adverse health effects in humans. Cypermethrin and deltamethrin are highly toxic to aquatic life, especially fish. The concentration of deltamethrin ranged from 27.3 to 227.5, 32.2-97.3 and 12.5-69.1 µg kg-1 for the River Makelele, Kalamu and Nsanga respectively. Since the year 2013, the Ministry of Agriculture of DRC has approved the import, sale, and use of deltamethrin and chlorpyrifos for insecticidal applications in agriculture (M.A.P., 2016). As a result, the above - mentioned chlorinated pesticides has been intensively used regularly in urban agriculture. Furthermore, many hospitals nearby urban river sites have been using insecticides for cleaning purposes. On other hand, the agricultural runoff and untreated hospital effluent waters are discharged into the river receiving systems without regulation. These aspects might have attributed to the high concentration of deltamethrin and other chlorinated pesticides at the studied sites, mainly in the site R1C (Makelele River) located near a great hospital discharging the effluent waters without previous treatment. OCPs including p,p'-DDE, o,p'-DDE, p,p'-DDD, o,p'-DDD, p,p'-DDT and o,p'-DDT were detected in all sediments as shown in Table 2-5. In all sediments, the most dominant isomers were in order p,p'-DDE > p,p'-DDD> p,p'-DDT > o,p'-DDT> o,p'-DDD and o,p'-DDE. The highest concentrations of Σ6 DDTs were recorded in the sediments from the Makelele River for site R1A (270.61 µg kg-1), whereas the lowest values detected at the site R3A (1.23 µg kg-1), Nsanga River. The fluctuations observed in the total concentrations of Σ6 DDTs suggest different emission sources on river basins or a dilution by clean sediments. The total concentrations of DDTs in all sediments was found to be higher than the TEL and PEL values except for the site R3A. A previous study reported that DDT could be converted into DDE by biodegradation under aerobic conditions via dehydrochlorination and oxidation process, and into DDD involving reductive dechlorination under anaerobic conditions (Syed et al. 2014). Similarly, our results showed elevated concentrations of DDE and DDD. The total concentrations of Σ6 DDTs detected in surface sediments in this study present an extreme increase in comparison to those previously detected in Congo River (Verhaert et al. 2013; Mwanamoki et al. 2014b). According to Li et al. (2016), it is possible to establish if DDTs input is historic or recent origin using the ratio between the concentrations of (DDD + DDE)/DDT. If the ratio is greater than 0.5 in the first case, indicates historic and less than 0.5 indicates recent input. The values of (DDD + DDE)/DDT ranged from 2.7 to 6.12, 2.2-8.9 and 2.2- 4.1 for the 59 River Makelele, Kalamu, and Nsanga, respectively. These results suggest that the rivers have been exposed to DDT from historic use, but may also be derived from indoor household appliances and agricultural misuse. DDTs are one of the important OCPs in connection with human and aquatic organism's health. According to WHO report, DDTs have been permitted for indoor household appliances in Congo DR for fighting against malaria vector (WHO, 2011). This could also possibly responsible for the elevated level of DDTs along the river basins. Σ6-PBDEs concentration ranged from 1.5-27.9 µg kg-1. BDE-47 and BDE-99 were the most abundant in all sites. We observe a concentration enrichment from upper stream to downstream in R1and R2 rivers. Figure 2-2c represent the results of PCA analysis for PBDEs. The PCA first component explained 98.3% of the total variance and was mainly influenced by contamination level. Only 1% of the total variance was explained by the second component. The different sites showed substantial differences between each other, highlighting the variability of PBDEs load and congener mixture in the sediment. Canadian Federal Environmental Quality Guidelines (FEQGs, 2013) recommended the threshold level of 44, 39, 0.4, 440, 5600 and 19 µg kg-1 dw in sediment for tri, tetra, penta, hexa, octa, and deca –BDEs respectively. The BDE99 concentrations at all sites exceed the recommended value of 0.4 µg kg-1 dw. The BDE100 concentrations were found higher than the threshold level of 0.4 µg kg-1 dw for most of the sediment samples. Coal combustion, urban sewage, the oil spill from the pirate garages and electronic wastes along the river basins were probably the possible sources of PBDEs in sediments. It should also be noted that the concentrations are much higher than in a previous work on Congo River Basin, DR Congo, where Verhaert et al. (2013) measured between below limits of quantification and 1.9 µg kg-1 dw for the sum of 8 BDEs, the same than in our case, more BDE-183 and BDE-209. BDE-209 represents near 90 % of Σ8-BDEs, followed by BDE-47 and BDE-99, 5 % and 3 % respectively. In another study on sediments from 6 rivers from Gauteng, South Africa, Olukunle et al. (2015) measured similar concentrations than in our study, values ranged from 0.82 µg kg-1 and 44 µg kg-1 for the sum of 8 PBDEs congeners, including BDE-209. Here, BDE-100, BDE-99 and BDE-47 were the most abundant congeners. In their study on 4 sampling sites from Murchison Bay of Lake Victoria, Uganda et al. (2012) measured concentrations ranged from 0.060 and 0.179 µg kg-1 for the sum of 11 PBDE congeners. The concentrations of individual PAHs investigated in the current study are presented in Table 2-4. All investigated PAHs were detected in most sediments. The concentrations of Σ16 PAHs in surface sediments ranged from 22.56-1011.94 µg kg-1. Based on PCA loading 60 plot (Figure 2-2d), we observed a strong correlation among all the congeners (R>0.83, p<0.01) except for naphthalene (0.40 < R< 0.74). The first component, which explains 93.3% of the total variance, was mainly influenced by the PAHs concentration in the sediment and the second component (4.9% of the total variance) was influenced by the PAHs mixture composition. PCA results showed that the composition of PAHs mixture differs substantially from site to site. The highest concentration for Σ16 PAHs was observed at R1C, while the lowest was recorded at R3A. Again, concentrations are higher downstream in R1 and R2 rivers, confirming the enrichment of PAHs across the more populated areas. As for ΣPCBs, ΣDDT, sampling point R1C on Makelele River is the most contaminated one. The results indicate that moderate to heavy contamination of sediments by PAHs. According to MacDonald et al. (2000b), proposed values for TEL were (1’610 µg kg-1) and PEL (22’800 µg kg-1) for the sum of 13 PAHs (the 16 PAHs analyzed in this work with exception of: dibenzo (a,h) anthracene, benzo (g,h,i) perylene and indeno (1,2,3c,d) pyrene). In general, the detection level of the sum of phenanthrene concentrations in all 9 sediments (635.24 µg kg-1) and of pyrene (550.36 µg kg-1) were found to be the most dominant PAHs among all sediments, followed by those of Naphthalene (457.86 µg kg-1) and those of fluoranthene (410.79 µg kg-1). The detection level of the sum of acenaphthylene in all sediments was the lowest (34.24 µg kg-1) among the 16 PAHs investigated in this study. As shown in Table 2-4, the high molecular weight (HMW, 4–6 aromatic rings) PAHs were predominant in surface sediments. The total concentrations of (Σ13 PAHs) in all sediments was found to be lower than the TEL value except for the sites R1C and R2B. However, the total concentrations of 13 PAHs in all sediments was found be lower than the PEL (22’800 µg kg-1 dw). Yang et al. (2013) reported that soils and sediments are the primary steady sinks for PAHs in the environmental compartments. A similar trend was observed in surface sediments of the current study. Furthermore, the results were in similar to that previously reported for rivers (Kanzari et al. 2014). However, it was noted that the concentrations of PAHs were higher than our previous study (Mwanamoki et al. 2014b) from the sediment of Congo River basins (34.48 to 63.89 µg kg-1). Yunker et al. (2002) reported that PAHs can originate from natural or anthropogenic processes. To better understand the potential source of PAHs, four diagnostic ratios were calculated in this study. According to Yunker et al. (2002) and Manneh et al. (1997), it is possible to establish if PAHs are from petrogenic or pyrogenic origin using the ratio between the concentrations of Fluo/(Fluo + Pyr) or IDP/(IDP + BghiP). If the ratio is < 0.4 in the first case, the source is petrogenic, when it is between 0.4 and 0.5 the source is petroleum 61 combustion, and when it is >0.5, the source is grass, wood, or coal combustion. The ratio of IDP/(IDP + BghiP) smaller than 0.2 indicates the source is petrogenic; a ratio ranged from 0.2 to 0.5 is considered as the source of petroleum combustion, a ratio greater than 5 indicates grass, wood, or coal combustion (Manneh et al. 2016). Furthermore, the ratio of BaA/(BaA+Chry) smaller than 0.2 is generally considered as petroleum source, a ratio between 0.2 and 0.35 indicates either a petroleum or combustion source and a ratio greater than 0.35 indicates pyrolytic origin (Manneh et al. 2016). In addition, Budzinski et al. (1997) propose that if the LMW/HMW ratio is smaller than 1, the source is petrogenic, when it is higher than 1, the source is pyrogenic. According to ratio as explained above, the results for sediment samples from Makelele and Nsanga Rivers suggested that the PAHs probably have originated from petrogenic processes and/or petroleum combustion (Table 2-6). The PAHs pollution in Kalamu River were probably originated from petrogenic and pyrogenic sources (combustion of fossil fuel). However, it should be noticed that the studied rivers take their sources upstream, in the savannas region, before arriving in the city. People leaving in the savannas burn the forest and the herbs for their crops without any regulation. Also in the city of Kinshasa, over 70% of the inhabitants use coal and wood to cook. This combustion of coal and wood could explain the origin of some PAHs along these rivers. Table 2-6 The values of Fluo/ (Fluo + Pyr), IDP/ (IDP + BghiP), BaA/ (BaA + Chry), and LMW/HMW ratios. Makelele Kalamu Nsanga Fluo/ (Fluo + Pyr) 0.39-0.46 0.46-0.48 0.41-0.42 IDP/ (IDP + BghiP) 0.39-0.41 0.42-0.44 *NA-0.46 BaA/ (BaA + Chry) 0.24-0.35 0.33-0.36 NA-0.40 **LMW/HMW 0.52-0.56 0.56-0.98 0.44-0.51 *Not available due to the concentration of individual PAH compound being below the Limit of quantification. **LMW/HMW: Ratio of low molecular weight (LMW) PAHs (i.e. Naphthalene (Naph), Acenaphthylene (Acy), Acenaphthene (Ace), Fluorene (Fl), Phenanthrene (Phen), Anthracene (Anthr) to high molecular weight (HMW) PAHs (i.e. fluoranthene (Fluo), pyrene (Pyr), benzo[a]anthracene (BaA), chrysene (Chry), benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), indeno[1,2,3-cd]pyrene (IDP), dibenz[a,h]anthracene (DahA) ,and benzo[g,h,i]perylene (BghiP). 62 For more than 45 years, PAHs, PCBs, OCPs, and PBDEs have been of great environmental concern. Consequently, several studies have been performed to assess their levels and potential risk in air, soil, and surface sediments of rivers and lakes from Europe and the United States (e.g. Degrendele., 2016; Kanzari et al. 2014; Thevenon et al. 2013; Desmet et al. 2012; Echols et al. 2012; Poté et al. 2008). In many cases, the levels of PCBs in sediments are higher than those detected in this study. However, the PAHs concentration presented in this study were higher than those observed in our previous study in sediments from Swiss Lakes (Pardos et al. 2004; Poté et al. 2008; Thevenon et al. 2013). 2.3.5 Correlation between parameters The Spearman’s rank-order correlation values are presented in Table 2-7. The results showed a significant positive correlation among the metals Zn, Pb, and Cd. Zn showed strong positive correlation with Cd (R = 0.917, p < 0.05, n = 9) and Pb (R = 0.9, p < 0.05, n = 9). Cd also displayed positive correlation with Pb (R = 0.733, p < 0.05, n = 9). These results indicated that these metals could have originated from common sources with a similar transport pathway (runoff and streams) (Poté et al. 2008; Haller et al. 2009). The negative correlation between total organic carbon and Cd indicating a possible multiple sources and transport pathway, which may be the discharge of domestic wastes from the urban population. On the whole, the results suggested that these toxic metals could have originated from multiple sources into the river receiving systems. Total organic matter displayed positive correlation with PAHs (R =0.733, p < 0.05, n = 9) and PBDEs (R =0.767, p < 0.05, n = 9) indicating that probably had a similar source such as from point source of dumping of urban waste and industrial activities on the river basins and combustion of coal, wood and fossil fuel. The negative correlative between grain size and PBDEs indicating that the diverse pollution source, possibly from electronic waste. PCBs showed strong positive correlation with TC, indicating PCBs were more likely come from urban surface runoff. In addition, DDTs showed strong positive correlation with TP, indicating that the DDTs accumulation in the sediments could be attributed to urban activities, pesticide abuse, and household runoff. Furthermore, a positive correlation was observed between PCBs and PAHs. This result suggested that PCBs and PAHs might have had common sources and transport pathway like an urban runoff. Consequently, it becomes clear that industrial wastewater, urban effluents, artisanal activities, dumping the huge amount of urban waste and combustion of coal, wood, and fossil fuel were the main contribution sources of toxic metals and micro pollutants in the studied areas. 63 Table 2-7 Spearman’s rank-order correlation of selected parametersa analyzed in the surface sediments Grain Zn Cd Pb TOM size TC TN TP PCBs PAHs DDTs PBDEs Cu 0.517 0.533 0.500 0.00 0.0833 -0.350 -0.0667 -0.250 -0.050 0.100 -0.333 0.183 Zn 0.917 0.900 -0.283 0.117 -0.467 0.183 0.367 -0.167 -0.300 0.267 -0.183 Cd 0.733 -0.167 0.250 -0.667 -0.117 0.117 -0.350 -0.400 0.200 -0.200 Pb -0.467 0.150 -0.183 0.467 0.467 0.0167 -0.200 0.233 -0.267 TOM -0.633 0.117 -0.517 -0.117 0.533 0.733 0.367 0.767 Grain size -0.367 0.0333 -0.0167 -0.633 -0.617 -0.350 -0.667 TC 0.300 0.183 0.733 0.500 0.0667 -0.0333 TN 0.567 0.3 0.050 0.050 -0.0667 TP 0.4 0.233 0.733 0.0333 PCBs 0.883 0.450 0.517 PAHs 0.367 0.783 DDTs 0.283 aParameters include toxic metals, median grain size, total organic matter, TC, TN, TP, PCBs, PAHs, DDTs and PBDEs [n = 9, statistically significant coefficients (p < 0.05) are in bold. 2.4 Conclusion This study provides an extensive data on heavy metals, POPs and PAHs contamination level in Makelele, Kalamu and Nsanga Rivers in Congo DR. The results indicated that four toxic heavy metals including Cu, Zn, Cd and Pb were predominantly detected in most of the sediments and the contamination level was higher than other Congo River basins. Makelele River is the most contaminated in comparison with two others investigated rivers in the present study. In addition, the concentration levels of PCBs, DDTs and PAHs were generally high in Makelele River and it warrants the important steps on limit the input sources to prevent further food web contamination. The PBDEs concentrations were higher in investigated rivers comparatively with some values detected in many rivers from Sub-Saharan Africa. It was also noted that PAHs represented the highest contributor of organic micro-pollutants pollution in these rivers. Our findings suggest that these rivers were heavily polluted and may pose a great risk to human health and aquatic environment. Further research in aquatic organisms such as fish from these rivers may provide more comprehensive information on contaminants with regard to the aquatic life and human health. Industrial effluents, untreated urban effluents, automobile exhaust, e-waste, improper incineration of urban waste in landfills, run off, petroleum combustion and pyrogenic activities could be the major contributors for the investigated contaminants for the studied sites. Our study will be useful for the establishment 64 of effective water management strategies in urban river ecosystems in Congo DR, which can be applied in similar aquatic environment. References Bartoli, G., S. Papa, E. Sagnella and A. Fioretto (2012). "Heavy metal content in sediments along the Calore river: Relationships with physical–chemical characteristics." Journal of Environmental Management 95, Supplement: S9-S14. Budzinski, H., I. Jones, J. Bellocq, C. Piérard and P. Garrigues (1997). "Evaluation of sediment contamination by polycyclic aromatic hydrocarbons in the Gironde estuary." Marine Chemistry 58(1): 85-97. Cui, S., Q. Fu, L. Guo, Y.-F. Li, T.-x. Li, W.-l. Ma, M. Wang and W.-l. Li (2016). "Spatial– temporal variation, possible source and ecological risk of PCBs in sediments from Songhua River, China: Effects of PCB elimination policy and reverse management framework." Marine Pollution Bulletin 106(1–2): 109-118. de Boer, J., C. Allchin, R. Law, B. Zegers and J. P. Boon (2001). "Method for the analysis of polybrominated diphenylethers in sediments and biota." TrAC Trends in Analytical Chemistry 20(10): 591-599. Devarajan, N., A. Laffite, P. Ngelikoto, V. Elongo, K. Prabakar, J. I. Mubedi, P. T. M. Piana, W. Wildi and J. Poté (2015). "Hospital and urban effluent waters as a source of accumulation of toxic metals in the sediment receiving system of the Cauvery River, Tiruchirappalli, Tamil Nadu, India." Environmental Science and Pollution Research 22(17): 12941-12950. Díez, S., S. Delgado, I. Aguilera, J. Astray, B. Pérez-Gómez, M. Torrent, J. Sunyer and J. M. Bayona (2009). "Prenatal and Early Childhood Exposure to Mercury and Methylmercury in Spain, a High-Fish-Consumer Country." Archives of Environmental Contamination and Toxicology 56(3): 615-622. Doong, R.-a., S.-h. Lee, C.-c. Lee, Y.-c. Sun and S.-c. Wu (2008). "Characterization and composition of heavy metals and persistent organic pollutants in water and estuarine sediments from Gao-ping River, Taiwan." Marine Pollution Bulletin 57(6–12): 846- 857. Dupré, B., J. Gaillardet, D. Rousseau and C. J. Allègre (1996). "Major and trace elements of river-borne material: The Congo Basin." Geochimica et Cosmochimica Acta 60(8): 1301-1321. 65 Ghrefat, H. and N. Yusuf (2006). "Assessing Mn, Fe, Cu, Zn, and Cd pollution in bottom sediments of Wadi Al-Arab Dam, Jordan." Chemosphere 65(11): 2114-2121. Haller, L., J. Poté, J.-L. Loizeau and W. Wildi (2009). "Distribution and survival of faecal indicator bacteria in the sediments of the Bay of Vidy, Lake Geneva, Switzerland." Ecological Indicators 9(3): 540-547. Hsu, L.-C., C.-Y. Huang, Y.-H. Chuang, H.-W. Chen, Y.-T. Chan, H. Y. Teah, T.-Y. Chen, C.- F. Chang, Y.-T. Liu and Y.-M. Tzou (2016). "Accumulation of heavy metals and trace elements in fluvial sediments received effluents from traditional and semiconductor industries." Scientific Reports 6: 34250. Huang, X., R. A. Hites, J. A. Foran, C. Hamilton, B. A. Knuth, S. J. Schwager and D. O. Carpenter (2006). "Consumption advisories for salmon based on risk of cancer and noncancer health effects." Environmental Research 101(2): 263-274. Jain, C. K., D. C. Singhal and M. K. Sharma (2005). "Metal Pollution Assessment of Sediment and Water in the River Hindon, India." Environmental Monitoring and Assessment 105(1): 193-207. Kanzari, F., A. D. Syakti, L. Asia, L. Malleret, A. Piram, G. Mille and P. Doumenq (2014). "Distributions and sources of persistent organic pollutants (aliphatic hydrocarbons, PAHs, PCBs and pesticides) in surface sediments of an industrialized urban river (Huveaune), France." Science of The Total Environment 478: 141-151. Kumar, B., S. Kumar and C. S. Sharma (2013). "Ecotoxicological Risk Assessment of Polychlorinated Biphenyls (PCBs) in Bank Sediments from along the Yamuna River in Delhi, India." Human and Ecological Risk Assessment: An International Journal 19(6): 1477-1487. Lin, C., M. He, Y. Zhou, W. Guo and Z. Yang (2007). "Distribution and contamination assessment of heavy metals in sediment of the Second Songhua River, China." Environmental Monitoring and Assessment 137(1): 329. Maanan, M., B. Zourarah, C. Carruesco, A. Aajjane and J. Naud (2004). "The distribution of heavy metals in the Sidi Moussa lagoon sediments (Atlantic Moroccan Coast)." Journal of African Earth Sciences 39(3–5): 473-483. Malik, R. N., F. Mehboob, U. Ali, A. Katsoyiannis, J. K. Schuster, C. Moeckel and K. C. Jones (2014). "Organo-halogenated contaminants (OHCs) in the sediments from the Soan River, Pakistan: OHCs(adsorbed TOC) burial flux, status and risk assessment." Science of The Total Environment 481: 343-351. 66 Manneh, R., C. Abi Ghanem, G. Khalaf, E. Najjar, B. El Khoury, A. Iaaly and H. El Zakhem (2016). "Analysis of polycyclic aromatic hydrocarbons (PAHs) in Lebanese surficial sediments: A focus on the regions of Tripoli, Jounieh, Dora, and Tyre." Marine Pollution Bulletin 110(1): 578-583. Mubedi, J. I., N. Devarajan, S. L. Faucheur, J. K. Mputu, E. K. Atibu, P. Sivalingam, K. Prabakar, P. T. Mpiana, W. Wildi and J. Poté (2013). "Effects of untreated hospital effluents on the accumulation of toxic metals in sediments of receiving system under tropical conditions: Case of South India and Democratic Republic of Congo." Chemosphere 93(6): 1070-1076. Mwanamoki, P. M., N. Devarajan, B. Niane, P. Ngelinkoto, F. Thevenon, J. W. Nlandu, P. T. Mpiana, K. Prabakar, J. I. Mubedi, C. G. Kabele, W. Wildi and J. Poté (2015). "Trace metal distributions in the sediments from river-reservoir systems: case of the Congo River and Lake Ma Vallée, Kinshasa (Democratic Republic of Congo)." Environmental Science and Pollution Research 22(1): 586-597. Mwanamoki, P. M., N. Devarajan, F. Thevenon, E. K. Atibu, J. B. Tshibanda, P. Ngelinkoto, P. T. Mpiana, K. Prabakar, J. I. Mubedi, C. G. Kabele, W. Wildi and J. Poté (2014). "Assessment of pathogenic bacteria in water and sediment from a water reservoir under tropical conditions (Lake Ma Vallée), Kinshasa Democratic Republic of Congo." Environmental Monitoring and Assessment 186(10): 6821-6830. Mwanamoki, P. M., N. Devarajan, F. Thevenon, N. Birane, L. F. de Alencastro, D. Grandjean, P. T. Mpiana, K. Prabakar, J. I. Mubedi, C. G. Kabele, W. Wildi and J. Poté (2014). "Trace metals and persistent organic pollutants in sediments from river-reservoir systems in Democratic Republic of Congo (DRC): Spatial distribution and potential ecotoxicological effects." Chemosphere 111: 485-492. Poté, J., L. Haller, J.-L. Loizeau, A. Garcia Bravo, V. Sastre and W. Wildi (2008). "Effects of a sewage treatment plant outlet pipe extension on the distribution of contaminants in the sediments of the Bay of Vidy, Lake Geneva, Switzerland." Bioresource Technology 99(15): 7122-7131. Raghunath, R., R. M. Tripathi, A. V. Kumar, A. P. Sathe, R. N. Khandekar and K. S. V. Nambi (1999). "Assessment of Pb, Cd, Cu, and Zn Exposures of 6- to 10-Year-Old Children in Mumbai." Environmental Research 80(3): 215-221. Sakan, S. M., D. S. Đorđević, D. D. Manojlović and P. S. Predrag (2009). "Assessment of heavy metal pollutants accumulation in the Tisza river sediments." Journal of Environmental Management 90(11): 3382-3390. 67 Singh, K. P., D. Mohan, V. K. Singh and A. Malik (2005). "Studies on distribution and fractionation of heavy metals in Gomti river sediments—a tributary of the Ganges, India." Journal of Hydrology 312(1–4): 14-27. Syed, J. H., R. N. Malik, J. Li, C. Chaemfa, G. Zhang and K. C. Jones (2014). "Status, distribution and ecological risk of organochlorines (OCs) in the surface sediments from the Ravi River, Pakistan." Science of The Total Environment 472: 204-211. Tamim, U., R. Khan, Y. N. Jolly, K. Fatema, S. Das, K. Naher, M. A. Islam, S. M. A. Islam and S. M. Hossain (2016). "Elemental distribution of metals in urban river sediments near an industrial effluent source." Chemosphere 155: 509-518. Thevenon, F., L. F. d. Alencastro, J.-L. Loizeau, T. Adatte, D. Grandjean, W. Wildi and J. Poté (2013). "A high-resolution historical sediment record of nutrients, trace elements and organochlorines (DDT and PCB) deposition in a drinking water reservoir (Lake Brêt, Switzerland) points at local and regional pollutant sources." Chemosphere 90(9): 2444-2452. Thevenon, F. and J. Poté (2012). "Water Pollution History of Switzerland Recorded by Sediments of the Large and Deep Perialpine Lakes Lucerne and Geneva." Water, Air, & Soil Pollution 223(9): 6157-6169. Thevenon, F., N. Regier, C. Benagli, M. Tonolla, T. Adatte, W. Wildi and J. Poté (2012). "Characterization of fecal indicator bacteria in sediments cores from the largest freshwater lake of Western Europe (Lake Geneva, Switzerland)." Ecotoxicology and Environmental Safety 78: 50-56. Tshibanda, J. B., N. Devarajan, N. Birane, P. M. Mwanamoki, E. K. Atibu, P. T. Mpiana, K. Prabakar, J. Mubedi Ilunga, W. Wildi and J. Poté (2014). "Microbiological and physicochemical characterization of water and sediment of an urban river: N’Djili River, Kinshasa, Democratic Republic of the Congo." Sustainability of Water Quality and Ecology 3–4: 47-54. Varol, M. (2011). "Assessment of heavy metal contamination in sediments of the Tigris River (Turkey) using pollution indices and multivariate statistical techniques." Journal of Hazardous Materials 195: 355-364. Verhaert, V., A. Covaci, S. Bouillon, K. Abrantes, D. Musibono, L. Bervoets, E. Verheyen and R. Blust (2013). "Baseline levels and trophic transfer of persistent organic pollutants in sediments and biota from the Congo River Basin (DR Congo)." Environment International 59: 290-302. 68 Yang, Y., L. A. Woodward, Q. X. Li and J. Wang (2014). "Concentrations, Source and Risk Assessment of Polycyclic Aromatic Hydrocarbons in Soils from Midway Atoll, North Pacific Ocean." PLOS ONE 9(1): e86441. Yunker, M. B., R. W. Macdonald, R. Vingarzan, R. H. Mitchell, D. Goyette and S. Sylvestre (2002). "PAHs in the Fraser River basin: a critical appraisal of PAH ratios as indicators of PAH source and composition." Organic Geochemistry 33(4): 489-515. Zahra, A., M. Z. Hashmi, R. N. Malik and Z. Ahmed (2014). "Enrichment and geo- accumulation of heavy metals and risk assessment of sediments of the Kurang Nallah—Feeding tributary of the Rawal Lake Reservoir, Pakistan." Science of The Total Environment 470–471: 925-933. Zhang, J. and C. L. Liu (2002). "Riverine Composition and Estuarine Geochemistry of Particulate Metals in China—Weathering Features, Anthropogenic Impact and Chemical Fluxes." Estuarine, Coastal and Shelf Science 54(6): 1051-1070. Zhang, Y., C.-S. Guo, J. Xu, Y.-Z. Tian, G.-L. Shi and Y.-C. Feng (2012). "Potential source contributions and risk assessment of PAHs in sediments from Taihu Lake, China: Comparison of three receptor models." Water Research 46(9): 3065-3073. 69 CHAPTER 3 High levels of faecal contamination in drinking groundwater and recreational water due to poor sanitation, in the sub-rural neighborhoods of Kinshasa, Democratic Republic of the Congo A similar version of this chapter was published under the following reference: Kayembe, J. M., F. Thevenon, A. Laffite, P. Sivalingam, P. Ngelinkoto, C. K. Mulaji, J.-P. Otamonga, J. I. Mubedi and J. Poté (2018). "High levels of faecal contamination in drinking groundwater and recreational water due to poor sanitation, in the sub-rural neighborhoods of Kinshasa, Democratic Republic of the Congo." International journal of hygiene and environmental health 221(3): 400-408. DOI: 10.1016/j.ijheh.2018.01.003 71 Abstract In many urban and peri-urban areas of developing countries, shallow wells and untreated water from urban rivers are used for domestic purposes, including drinking water supply, population bathing and irrigation for urban agriculture. The evaluation and monitoring of water quality are therefore necessary for preventing potential human risk associated with the exposure to contaminated water. In this study, physicochemical and bacteriological parameters were assessed in an urban river (named Kokolo Canal/Jerusalem River) draining the municipality of Lingwala (City of Kinshasa, Democratic Republic of the Congo) and in two shallow wells used as drinking water supplies, during the wet and dry seasons in order to estimate the seasonal variation of contamination. The faecal indicator bacteria (FIB) isolated strains (Escherichia coli (E. coli) and Enterococcus (ENT)) from water and surface sediment, were characterized for human-specific Bacteroides by molecular approach. The results revealed very high faecal contamination of water from the shallow wells, and of water and sediments from the river, during both wet and dry seasons. During the wet season, E. coli reached the values of 18.6x105 and 4.9x105 CFU 100 mL-1 in Kokolo Canal and shallow wells, respectively; and Enterococcus reached the values of 7.4x104 and 2.7x104 CFU 100 mL-1. The water samples from the shallow wells and Kokolo Canal were highly polluted with faecal matter in both seasons. However, the pollution level was significantly higher during the wet season compared to the dry season. Physicochemical analysis revealed also very high water electrical conductivity, with values much higher than the recommended limits of the World Health Organization guideline for drinking water. These results highlight the potential human health risk associated with the exposure to water contamination from shallow wells and Kokolo Canal, due to the very high level of human FIB. Rapid, unplanned and uncontrolled population growth in the city of Kinshasa is increasing considerably the water demand, whereas there is a dramatic lack of appropriate sanitation and wastewater facilities, as well as of faecal sludge (and solid waste) management and treatment. The lack of hygiene and the practice of open defecation is leading to the degradation of water quality, consequently the persistence of waterborne diseases in the neighbourhoods of sub-rural municipalities, and there is a growing threat to the sustainability to water resources and water quality. The results of this study should encourage municipality policy and strategy on increasing the access to safely managed sanitation services; in order to better protect surface water and groundwater sources, and limit the proliferation of epidemics touching regularly the city. 72 3.1 Introduction The 2030 Agenda which was one of the outcomes from the Rio +20 conference of 2012, and the adoption of Sustainable Development Goals (SDGs) by all member states of the United Nations in 2015, dedicated a goal to water and sanitation (SDG 6) with its first Target (6.1) focusing on the universal and equitable access to safe and affordable drinking water for all (UN- Water, 2016). There is also an urgent need for increasing the access to adequate and equitable sanitation and hygiene and end open defecation (Target 6.2), not only for human dignity, but also for protecting the quality of natural drinking water sources that are frequently contaminated by faecal pathogens. In order to improve the ambient water quality, which is essential to protect human health (and ecosystem health), the SDG 6 framework entails halving the proportion of wastewater generated by households (and all economic activities) that is untreated, and substantially increasing recycling and safe reuse globally (Target 6.3). There is however, a major concern with respect to the quality of drinking water in rapidly developing mega-cities of low and middle-income countries, where people are drinking untreated surface water and groundwater. Many urban rivers are heavily polluted due to the large discharge of untreated domestic, hospital and industrial effluents, the frequent presence of landfills near the river banks, and poultry farming manure (Laffite et al. 2016). Moreover, the data concerning the occurrence of pathogenic organisms in these aquatic environments are very limited (Rochelle-Newall et al. 2015; Rodriguez-Alvareza et al. 2015; Nienie et al. 2017). Due to the economic situation and to the lack of effective water treatment infrastructure in many regions of developing countries, peoples are directly using contaminated water from rivers, shallow wells, boreholes, springs and streams for irrigation, domestic and drinking purposes; which has the potential to significantly impact human health (Mubedi et al. 2013; Tshibanda et al. 2014; Mwanamoki et al. 2015; Kapembo et al. 2016; Laffite et al. 2016; Martínez-Santos et al. 2018). According to the WHO/UNICEF Joint Monitoring Program (JMP) for Water Supply, Sanitation and Hygiene (WHO/UNICEF, 2017) which is used to monitor international progress in access to drinking water, sanitation and hygiene, in 2015, i) 29% of the global population (2.1 billion people) lacked “safely managed drinking water” (meaning water at home, available, and safe); and ii) 61% of the global population (4.5 billion people) lacked “safely managed sanitation” (meaning access to a toilet or latrine that leads to treatment or safe disposal of excreta). In the Democratic Republic of the Congo (DRC), with an estimated population of 73 about 70 million inhabitants and despite the potential of its rich freshwater resources, more than 75% of the population have no access to in-home piped water (UNEP, 2011). At a national level, Water, Sanitation and Hygiene (WASH) access in DRC lags behind most of sub-Saharan Africa, with 47% of the population using unimproved sanitation facilities, 36% of the population using unimproved drinking water services, and 84% of the population using no hygiene facilities (i.e., handwashing facilities with soap and water) (WHO/UNICEF, 2017). Moreover, 16% of the rural population is using surface water as drinking water source (i.e., rivers, streams, wells, and springs). The evaluation of water quality from natural sources is still very limited in DRC due to the absence of national and state's water quality monitoring system. The evaluation of the microbiological quality of natural water (rivers and lakes) can be performed in both water and surface sediments, in order to assess seasonal and inter-annual contamination trends, respectively (Alm et al. 2003; Craig et al. 2004; Lee et al. 2006; Poté et al. 2009). The FIB including E. coli (a subset of the faecal coliform group) and members of the genus Enterococcus (the enterococci (ENT)), are recommended by the US Environmental Protection Agency, the European Environment Agency and the World Health Organization to assess the contamination of water with faecal matter (Haller et al. 2009; Brinkmeyer et al. 2015). The main objective of this research is to assess the seasonal variation of water quality from the Kokolo Canal (KC) and from two shallow wells. The KC is one of important river sources of Kinshasa City for urban agricultural irrigation, recreational activities, fishing and domestic purposes. It receives the untreated urban and hospital effluent waters and serves as sources of domestic purpose, including bathing, but also irrigation for urban agriculture. The shallow wells located at the vicinity of KC are serving for domestic use including drinking, cooking and washing. The water quality assessment is based on (i) the quantification of water physicochemical parameters including pH, electrical conductivity (EC) and dissolved oxygen (O2), (ii) the quantification of FIB including E. coli and ENT in water and sediment, and (iii) the characterization of the isolated FIB strains by molecular approach, in order to identify the source of water contamination. 74 3.2 Materials and Methods 3.2.1 Study site description This study was performed in an important urban river, the Kokolo Canal (KC) named Jerusalem River and two shallow wells (P1) and P2, both located near the river. KC is flowing through the urban commune of Lingwala and traversing the military region (named Camp Kokolo) in Kinshasa. Kinshasa is the capital and largest city of the Democratic Republic of the Congo (DRC) (Figure 3-1 Location map of the study area. A: Location Map of Congo DR in Africa. B: Location Map of Kinshasa City in Congo DR. C: Location map of studied River (Kokolo Canal) and shallow wells P1 and P2 at Kinshasa, Congo DR.), with an estimated population of about 12 million inhabitants and covering 9’965 km². Kinshasa has a tropical wet and dry climate. Its length rainy season spans from October through May, and a relatively short dry season occurs between June and September. Daily temperature varies between 25-38oC (during wet season) and 12-18oC (during dry season). The average annual rainfall is about 124 mm, but it often varies between 2 mm and 250 mm. Figure 3-1 Location map of the study area. A: Location Map of Congo DR in Africa. B: Location Map of Kinshasa City in Congo DR. C: Location map of studied River (Kokolo Canal) and shallow wells P1 and P2 at Kinshasa, Congo DR. 75 3.2.2 Sampling procedure Water (n=3) and sediment (n=3) samples from KC were collected at the same points in April 2017 (wet season) and in July 2017 (dry season). KC water samples were collected manually at 10-50 cm water depth and about 50 cm from the shore, and labelled as KCW1- KCW3. Water from shallow wells (n=2) was taken by a craft device made of 1 L clean polyethylene bottle attached to a rope (Kapembo et al. 2016), and labelled P1 and P2. The water samples (500 mL sealed in clean plastic bottles) were collected in triplicate from each sampling site. While sampling water at each site, three clean plastic bottles of 2 L containing Milli-Q water were kept open to the air to estimate field controls (Nienie et al. 2017a). The KC sediment samples (0-5 cm) were collected at a distance less than 50 cm from the shore and at less than 50 cm water depth. From each sampling point, about 350 g of sediment samples were collected in triplicate using a sterile plastic spoon and transferred into 1.5 L sterile bottles. The sediment samples are labelled KCS1-KCS3. Once taken, the samples were stored in an icebox and transported to the laboratory for analysis within 48 h. The sampling points in the river as well as for the shallow well were selected according to their frequent use for domestic and agricultural purposes (Figure 3-2 and Figure 3-3). A referential hospital rejecting untreated effluent water into KC (HOP) has been taken as reference in order to evaluate the possible contribution of this hospital effluent to the river contamination by FIB. For this reason, the water and sediment samples have been taken upstream and downstream to the reject point, and in the hospital effluent outlet pipe (HOP). The samples from HOP are labelled KCW4 and KCS4 for water and sediment, respectively. Sample description and GPS geographical coordinates of sampling sites are presented in Table 3-1. Table 3-1 GPS Location and description of sampling sites Sampling sites Sample name Comment Latitude Longitude KCW1/KCS1 Upstream 04°20’00.8’’S 015°17’24.6’’E KCW2/KCS2 Reject point 04°20’00.2’’S 015°17’23.1’’E Kokolo Canal KCW3/KCS3 Downstream 04°19’59.6’’S 015°17’21.6’’E HOP KCW4/KCS4 Hospital outlet pipe 04°20’01.4’’S 015°17’23.4’’E 04°20'01.7'' S 015°17'23.9'' E P1 About 1000 users Shallow well 04°20’01.8’’S 015°17’23.0’’E P2 About 1300 users KCW1-3 and KCS1-3: Water and sediment samples, respectively from Kokolo Canal KCW4 and KCS4: water and sediment samples, respectively from hospital outlet pipe (HOP) discharging to the Kokolo Canal. P1 and P2: Samples from shallow well 76 Figure 3-2 Photos taken by John Kayembe in April 2017. A: Bathing, recreational activity in Kokolo Canal; B: Kokolo canal using for domestic purpose; C: Shallow well P1 using for domestic purpose; D: Shallow well P2 using for domestic and agricultural purpose Figure 3-3 Photos taken by John Kayembe in July 2017. A: Pit latrines located in the bank of Kokolo Canal; B: Recreational activity (children playing foot) near Kokolo Canal; C: Children playing in Kokolo canal; D: Children bathing/enjoying in Kokolo Canal after playing foot 77 3.2.3 Water physicochemical parameter analysis Water Physicochemical parameters including temperature (T), pH, dissolved oxygen (O2) and electrical conductivity (EC) were measured in situ using a Multi parameter 350i (WTW, Germany). 3.2.4 Faecal indicator bacteria (FIB) analysis in water and sediment samples The FIB (including E. coli and ENT) were quantified in water samples and sediment supernatants according to the international standard methods for water quality determination using the membrane filtration method (APHA 2005). E. coli and ENT analysis in sediment samples was performed as described by Haller et al. (2009); Poté et al. (2009) and Kilunga et al. (2016). Briefly; the sediments were resuspended by adding 100 g of fresh sediment to 500 mL of 0.2 % Na6(PO3)6 in 1 L sterile plastic bottles and mixed for 30 min using the agitator rotary printing-press Watson-Marlow 601 controller (Skan, Switzerland). The mixture was then centrifuged at 4000 rpm (Sigma, 3-16K) for 15 min at 15°C. For each sample, triplicates of serially diluted sediment supernatant (100 mL) were used. Water samples and sediment supernatant were then passed through a 0.45 µm filter (47 mm diameter, Millipore, Bedford, USA), and placed on different selective culture media (Biolife Italiana, Milano Italy) supplemented with the anti-fungal compound Nystatin (100 g mL-1 final concentration), using the following incubation conditions: E. coli: Tryptone Soy Agar (TSA) medium, incubated at 37°C for 4 h and transferred to Tryptone Bile X-Gluc Agar (TBX) medium at 44°C for 24 h; ENT: Slanetz Bartley Agar (SBA) medium, incubated at 44°C for 48 h and transferred into Bile Aesculin Agar (BAA) medium at 44°C for 4 h. The results were expressed as colony forming units per 100 mL of water (CFU 100 mL-1) or 100 g of fresh sediments (CFU 100 g-1). The reproducibility of the whole experimental procedure was estimated by means of triplicates on selected sediment samples. The sample revealed a mean variation coefficient of 7 and 9% for E. coli and ENT respectively. 3.2.5 Characterization of FIB strains The characterization of FIB isolated strains was performed as described by Thevenon et al. (2012) and Nienie et al. (2017). Briefly; before PCR human-specific Bacteroides 78 amplification, the genomic profiles of general origin of E. coli and ENT were performed by PCR assays (presence/absence) using specific primers and operational conditions as summarized in Table 3-2 (Ahmed et al. 2007; Ke et al. 1999; Berhard et al. 2000; Sabat, Morrison et al. 2008 ; Scott et al. 2005). More than 500 isolated colonies from KC (water and sediment samples), P1 and P2 were selected. PCR assays to confirm the human-specific Bacteroides were performed as previously described by Thevenon et al. (2012); Tshibanda et al. (2014) and Kilunga et al. (2016) using human-specific Bacteroides primers shown in Table 3-2. The experiment was conducted in triplicate for each set of conditions. The negative (without DNA) and positive controls (e.g. 520 bp length expected for HF183/Bac708 from sewage (Poté et al. 2009) were used for each PCR assay. Table 3-2 Primers used for PCR amplification of general E. coli and Enterococci, and human- specific bacteroides* Size Anealing Primers Target PCR Sequence (5’ to 3’) Reference T°(°C) prod. ECA75F GGAAGAAGCTTGCTTCTTTGCTGAC General E. coli 544 60 Sabat et al. 2000 ECA619R AGCCCGGGGATTTCACATCTGACTTA Ent1 General 112 TACTGACAAACCATTCATGATG 55 / 49 Ke et al. 1999 / Ent2 Enterococci AACTTCGTCACCAACGCGAAC Morrison et al. 2008 HF183/134 human HF183 520 ATCATGAGTTCACATGTCCG human HF134 ATCARGTCACATGTCCCG Bernhard and Field, 59 2000 / Ahmed et al. Bac708R CAATCGGAGTTCTTCGTG 2007 570 ATCARGTCACATGTCCCG *The operational conditions for PCR amplification were carried out according to the published methods (references in this Table with minor modification Thevenon et al. (2012);Tshibanda et al. (2014); Kilunga et al. (2016)). 3.2.6 Data Analysis All analyses were conducted in triplicate for each set of conditions. In addition, three plates per dilution were performed for FIB quantification to establish plate count standard deviation. Statistical processing of data (Spearman's Rank-Order Correlation) was performed using SigmaStat 11.0 (Systat Software, Inc., USA). 79 3.3 Results and discussion 3.3.1 Water physicochemical parameters The results of water physicochemical parameters including T, pH, EC, and O2 according to the seasonal variation are presented in Table 3-3. In Kokolo Canal and HOP, the values during the wet season ranged from 27.4-28.8°C (T), 7.3-7.4 (pH), 323.0-614.0 µS cm-1 (EC) -1 and 0.2-0.6 mg L (O2) while dry season values ranged from 25.3-27.4°C (T), 6.6-7.4 (pH), -1 -1 323-567 µS cm (EC) and 0.2-0.9 mg L (O2). Except for electrical conductivity, there was no significant difference in pH, T and O2 levels according to the seasonal variation (p˃0.05). The EC values varied significantly according to the seasonal variation and sampling sites in KC (p ˂ 0.05). The physicochemical results observed in KC are comparable with our previous studies performed in the region (Mubedi et al. 2013; Tshibanda et al. 2014). Table 3-3 Physicochemical parameters of water samples from Kokolo Canal and wells (P1 and P2) during the dry season (dry) and wet season (wet) Sampling sites Sample name T°C pH Cond. (µS cm-1) O2 (mg L-1) Wet Dry Wet Dry Wet Dry Wet Dry KCW1 27.4 27.4 7.4 7.4 323.0 323.0 0.5 0.5 Kokolo Canal KCW2 28.0 26.0 7.3 6.8 401.0 458.0 0.3 0.2 KCW3 28.8 26.4 7.3 7.1 533.0 567.0 0.2 0.3 HOP KCW4 28.2 25.3 7.3 6.6 614.0 325.0 0.6 0.9 Minimum 27.4 25.3 7.3 6.6 323.0 323.0 0.2 0.3 Maximum 28.8 27.4 7.4 7.4 614.0 567.0 0.6 0.9 Shallow wells P1 28.2 25.8 6.4 5.7 543.0 527.0 1.8 3.3 P2 27.8 25.3 6.9 6.1 552.0 568.0 2.1 2.9 In shallow wells, during the wet season, the average values of physicochemical parameters ranged from 27.8-28.2°C (T), 6.4-6.9 (pH), 543.0-552.0 µS cm-1 ( EC), 1.8-2.1 mg -1 -1 L (O2) while the values of 25.3-25.8°C (T), 5.7- 6.1 (pH), 527.0-568.0 µS cm (EC) and 2.9- 3.3 mg L-1 (O2) were observed during the dry season. Except for the EC, the values of physicochemical parameters in P1 and P2 comply within WHO recommendation for drinking water (WHO, 2011) and are comparable with other published data obtained under tropical conditions (Pritchard et al. 2008; Nola et al. 2013; Nienie et al. 2017a). In P1, the EC reach the values of 543 and 527 µS cm-1 during the wet and dry season, respectively, while in P2 the value of 552 and 568 µS cm-1 were observed during the wet and dry season, respectively. These 80 EC values in P1 and P2 are higher than the limit recommended by WHO. However, the EC values from P1 and P2 are lower than the values found by Kapembo et al. (2016), showing the high values of conductivity in shallow wells from municipality of Bumbu, city of Kinshasa. The authors found the values ranging from 618-1547 and from 605-1121 µS cm-1 during the dry and wet season, respectively. 3.3.2 River microbiological quality The variation of faecal indicator bacteria (FIB) levels in water and sediment samples from Kokolo Canal (KC) are presented in Figure 3-4. Water samples from all sampled sites showed highly level of FIB which varied significantly according to sampling sites and seasonal variation (P ˂ 0.05). The FIB levels in water samples from the KC recorded during the wet season ranged from (1.8-18.6) x 105 and (1.3-7.4) x 104 CFU 100 mL-1 for E. coli and ENT, respectively. On other hand, the FIB values recorded in the dry season ranged from (1.1-6.8) x 105 and (0.9-4.7) x 104 CFU 100 mL-1 for E. coli and ENT, respectively. These results indicate that studied river is substantially polluted with faecal matter with FIB concentrations exceeded the limits for bathing water according to both European Directive (EU, Directive 2006/7/CE) and WHO guidelines for domestic use water. According to the European Directive 2006/7/CE concerning the management of bathing water quality, recreational waters are to be classified as poor, if concentrations of E. coli exceed 900 CFU·100 mL-1 and concentrations of ENT exceed 330 CFU 100 mL-1, based upon a 90 -percentile evaluation (Haller et al. 2009). By comparison, the FIB in KC water largely exceeded the legal limits for the microbial quality and the contamination of bathing waters according to the WHO regulation. 81 E. Coli in water samples Wet Season ENT in water samples Wet Season E. Coli in water samples Dry Season ENT in water samples Dry Season 1 - 1 - L L m m 0 20 0 0 0 1 1 5 8 4 0 0 1 1 x x ) ) D D S S ± ± U U F F C C ( ( i l T o N C E . E 10 4 0 0 1 2 3 1 2 1 2 3 4 l 1 2 4 l P P a P P W W W W ita W W W W it C C C C C C C p K K K C sp K K K s K o K o H H Kokolo Canal Shallow wells Kokolo Canal Shallow wells E. Coli in sediment Wet Season ENT in sediment Wet Season E. Coli in sediment Dry Season ENT in sediment Dry Season 80 60 1 - g 1 0 - 0 g 1 0 5 60 0 0 1 1 5 x 0 ) 1 x D ) S D ± S ± U 40 U F F C C ( ( i l 40 T o N C E . E 20 20 0 0 l 1 2 3 4 l 1 2 3 4 a S S S S ita S S S S it C C C p C C C C p K K K C s K K K s K o K o H H Kokolo Canal Kokolo Canal Figure 3-4 Average of Escherichia coli and Enterococcus quantification in water and sediment samples from river (Kokolo Canal), Hospital Outlet Pipe and shallow wells (P1 and P2) during the wet and dry seasons* * KCW1-3 and KCS1-3: Water and sediment samples, respectively from Kokolo Canal; KCW4 and KCS4: water and sediment samples, respectively from hospital outlet pipe discharging to the Kokolo Canal; P1 and P2: Water samples from shallow wells. 82 The sediments from KC present relatively similar high concentrations of FIB during both, wet and dry seasons, because the microbiological (and geochemical) composition of the surface sediments does not vary quickly like the composition of water – but rather represents the inter-annual composition of the water. The FIB values in sediment samples from KC recorded during the wet season ranged from (1.1-33.2) x 105 and (8.4-59.7) x 105 CFU 100 g-1 for E. coli and ENT, respectively. In addition, the FIB concentrations in sediment samples recorded in the dry season ranged from (2.4-62.8) x 105 and (7.7-53.4) x 105 CFU 100 g-1 for E. coli and ENT, respectively (Table 3-4). Although there are no health standards to evaluate the contamination of sediment by FIB, according to several researches, the present level of contamination of river sediments can be considered as very preoccupants. These results also indicate that the single analysis of water quality could underestimate the risk of exposure to potentially pathogenic microorganisms in recreational waters (Craig et al. 2002; Lee et al. 2006; Poté et al. 2009). It has been furthermore demonstrated that sediments may contain 100 to 1000 times as many FIB as the overlying water, and FIB can survive longer in sediments than in the water column, as sediments provide favourable conditions for their proliferation and growth (Davies et al. 1995; Poté et al. 2009). Resuspension of FIB and pathogens from the sediments to the water column can occur during the recreational activities and may contribute to potential human health risk (Craig et al. 2004; Haller et al. 2009). Therefore, the high levels of FIB found in the KC sediments may cause considerable water failures and children risks during bathing (Figure 3-2A). Water and sediment samples from the hospital outlet pipe (HOP) discharging to the Kokolo Canal present high concentration of FIB during both, wet and dry season. Concerning the water, during the wet season, the values ranged from (1.7-2.6) x 105 and (1.3-3.7) x 105 CFU 100 mL-1 for E. coli and ENT respectively. During the dry season, the values ranged from (0.4-2.1) x 105 and (0.9-2.5) x 105 CFU 100 mL-1 for E. coli and ENT respectively. Additionally, in sediment samples during the wet season, the values ranged from (7.5-12.4) x 105 and (12.4- 48.4) x 105 CFU 100 g-1 for E. coli and ENT respectively. During the dry season, the values ranged from (1.8-14.8) x 105 and (8.8-39.6) x 105 CFU 100 g-1 for E. coli and ENT respectively (Table 3-4). These values are lowers than those found in water and sediment upstream of the river, indicating that the hospital effluent cannot be considered as an exclusive source of river contamination by faecal material. The urban canal pollution could be rather explained by possible multiple sources, especially inadequate sanitation and open defecation practices, and urban and agricultural runoff (Mwanamoki et al. 2015; Laffite et al. 2016; Kilunga et al. 2016). 83 Table 3-4 Average of Escherichia coli and Enterococcus quantification in water and sediment samples from river (Kokolo Canal), hospital outlet pipe and shallow wells (drinking water) during the wet and dry seasons Watera Sediment E. coli ENT E. coli ENT Sample* (CFU± SD)x105 100 m L-1 (CFU± SD)x104 100 m L-1 (CFU ± SD)x105100 g-1 (CFU± SD)x105100 g-1 Wet Dry Wet Dry Wet Dry Wet Dry KCW1/KCS1 Min. 3.2 ± 1.8 1.4 ± 0.9 2.9 ± 1.4 1.7 ± 0.5 1.1 ± 0.7 2.4 ± 1.3 8.4 ± 2.5 7.7 ± 1.4 Max. 9.5± 2.6 3.7 ± 1.8 3.8 ± 1.2 2.6 ± 0.9 33.2 ± 12.4 41.5 ± 12.3 56.0 ± 16.7 44.0 ± 15.8 KCW2/KCS2 Min. 1.8 ± 0.7 1.1 ± 0.7 1.7 ± 0.6 0.9 ± 0.2 0.2 ± 0.1 3.7 ± 0.3 16.0 ± 4.8 12.0 ± 1.9 Max. 5.7 ± 0.3 2.6 ± 1.2 4.2 ± 1.8 2.7 ± 1.2 27.5 ± 6.7 33.4 ± 7.8 45.0 ± 11.9 49.0 ± 13.2 KCW3/KCS3 Min. 4.5 ± 0.4 3.9 ± 01.3 2.8 ± 1.3 1.8 ± 1.1 0.9 ± 0.6 2.1 ± 0.7 14.1 ± 3.6 13.2 ± 4.1 Max. 18.6 ± 5.8 6.8 ± 1.9 7.4 ± 2.6 4.7 ± 1.5 0.3 ± 0.1 62.8 ± 16.8 59.7 ± 17.2 53.4 ± 15.5 KCW4/KCS4 Min. 1.7 ± 0.7 0.4 ± 0.1 1.3 ± 0.8 0.9 ± 0.2 7.5 ± 1.3 1.8 ± 0.5 12.4 ± 4.1 8.8 ± 2.2 Max. 2.6 ± 1.3 2.1 ± 0.5 3.7 ± 1.7 2.5 ± 0.4 12.4 ± 2.6 14.8 ± 3.6 48.4 ± 13.8 39.6 ± 7.4 P1 Min. 0.3 ± 0.1 0.2 ± 0.1 0.3 ± 0.1 0.3 ± 0.1 n.a n.a n.a n.a Max. 1.8 ± 1.1 0.7 ± 0.2 0.6 ± 0.2 0.5 ± 0.3 n.a n.a n.a n.a P2 Min. 0.5 ± 0.2 0.8 ± 0.2 0.3 ± 0.2 0.2 ± 0.1 n.a n.a n.a n.a Max. 4.9 ± 0.6 1.5 ± 0.2 2.7 ± 0.5 0.8 ± 0.3 n.a n.a n.a n.a 3.3.3 Microbiological quality of water from wells The microbial analysis of the investigated shallow wells showed a high contamination of water by faecal material during both dry and wet season (Table 3-4). During the wet season, the FIB values ranged from (0.3-4.9) x105 and (0.3-2.7) x105 CFU 100 mL-1 for E. coli and ENT, respectively; while during dry season the values ranged from (0.2-1.5) x105 and (0.2-0.8) x105 CFU 100 mL-1 for E. coli and ENT, respectively. Interestingly and as previously suggested by the authors in a similar environment, the results of this study showed that the concentration of E. coli and ENT in shallow wells during the wet season could increase by 2 to 3 orders of magnitude compared to those measured during the dry season (Kapembo et al. 2016; Kilunga et al. 2016; Nienie et al. 2017). Such increasing contamination by faecal bacteria during the wet season may result from the higher runoff and from the overflow of onsite sanitation systems (e.g. pit latrines and septic tanks) into urban drains. The higher concentration of FIB in water indicate the potential presence of pathogenic organisms responsible for water-related diseases such as gastro-intestinal illnesses, typhoid, cholera, and other diarrhoeal diseases (Noble et al. 84 2004; Davis et al. 2005). These diseases are recurrent and persistent in the peri-urban communes of Kinshasa City which are characterized by low economic status (Kapembo et al. 2016). As demonstrated in our previous studies performed in similar context (Nienie et al. 2017; Kilunga et al. 2016), the pollution of KC and in the investigated wells can be attributed to several potential sources including the lack of hygiene, inadequate sanitation, the lack of safe management of the human waste, the non-protection of wells, the presence of pit latrines located in the proximity of wells, and open defecation practice (Figure 3-2C,D and Figure 3-3A). As a consequence, the precipitation poses important challenges for Kinshasa’s surface drinking water safety, since the frequency and the intensity of floods is increasing the contamination of natural water sources by excreta, thus creating additional health risks for the urban population. The results showed that the concentration of FIBs in rivers and wells are higher than 1000 times US EPA recommended concentrations in irrigation water (US EPA, 2004). Intriguingly, the results appear to be a strong recommendation for the further epidemiological study to define the probabilistic dose-effect relationship. 3.3.4 Characterization of Faecal Indicator Bacteria The FIB characterization is very important in order to assess the water contamination by human material and the prevention of human health risk for drinking water or during the recreational activities (Scott et al. 2005; Converse et al. 2009). In this study, qualitative PCR assays was performed for screening of the FIB isolated strains from water and sediment samples. The results are presented in Table 3-5. More than 500 isolated strains of E. coli and ENT were screened by PCR amplification using the primers HF183/HF134 for faecal human pollution (human-specific Bacteroides) as performed in our previous studies (Thevenon et al. 2012; Tshibanda et al. 2014). The results showed that more than 98% of FIB isolated from sediment samples and 100% of strains from water samples in both the wet and dry season seasons were of human origin, consequently the potential human health risks associated with direct use of water from wells and use of river during the recreational activities. The same results were observed in the samples from HOP. Such results should encourage the local authorities to conduct an epidemiological survey in selected exposure groups, in order to further limit the proliferation of epidemics touching regularly the city. Multi stakeholder workshops and awareness campaigns are also recommended, in order to support the government's long- term strategy to substantially increase the access to safely managed sanitation services; as well 85 as to improve the protection of drinking/recreational water sources and the health of crop/vegetable producers and consumers in the peri-urban areas of Kinshasa. Table 3-5 PCR presence/absence assays for detection of E.coli and Enterococcus in water samples from wells, rivers, hospital outlet pipes, and sediment samples from rivers (Kokolo Canal). Sampling sites Sample name Escherichia coli Enterococcus Wet Dry Wet Dry NT NP NT NP NT NP NT NP KCW1 15 15 14 14 12 12 9 9 KCW2 15 15 14 14 12 12 9 9 KCW3 15 15 14 14 12 12 9 9 Kokolo Canal KCS1 17 14 13 13 14 14 9 9 KCS2 17 16 13 11 14 14 9 6 KCS3 17 17 13 13 14 12 9 8 Shallow wells P1 28 28 18 18 10 10 10 10 P2 26 26 18 18 10 10 10 10 Hospital out let KCW4 22 22 14 14 n.a n.a n.a n.a pipe KCS4 16 13 14 8 n.a n.a n.a n.a NT: number of tested colonies, NP: number of positive PCR, n.a: analysis not performed 3.3.5 Correlation between parameters The results of Spearman's Rank-order correlation of data from Kokolo Canal are presented in Table 3-6. The strong positive mutual correlation (p-value < 0.005) was observed between pH, T, EC, E. coli and ENT in water samples from the KC. The same trend was observed in water samples from P1 and P2 during the wet season showing a strong mutually positive correlation between E. coli and ENT with the range of R-value being 0.93 < R < 0.97 (p-value < 0.001, n=15). Additionally, water analysis collected from the P1 and P2 in the dry season shown E. coli and ENT with a mutually positive with R-values ranged from 0.87 to 0.98 ( p-value < 0.05, n=15). These results indicate that during both, wet and dry season, the FIB in water samples could be considered to originate from common sources and they are carried to the river receiving system as well as in shallow wells by common transporters (Haller et al. 2009; Kilunga et al. 2016; Kapembo et al. 2016). 86 Table 3-6 Spearman's Rank-Order Correlation of selected parameters* analysed in water samples from Kokolo Canal. pH E. coli ENT T O2 EC 0.837 0.792 0.867 0.691 0.283 pH 0.768 0.973 0.764 0.519 E. coli 0.988 0.836 0.735 ENT 0.763 0.834 T 0.916 * Analysed parameters include Temperature (T), pH, Electrical conductivity (EC), dissolved oxygen (O2), Escherichia coli (E. coli) and Enterococcus (ENT). Significant coefficients (p < 0.05) are in bold. 3.4 Conclusion Microbiological analyses of water and sediments samples from shallow wells and Kokolo Canal during the wet season showed that the wells and river are substantially polluted with faecal matter and do not meet the WHO guideline for bathing, drinking and domestic purpose. Interestingly, the highly FIB values observed in water and sediment samples located upstream of the hospital outlet discharges indicate that this hospital cannot be considered as a unique source of deterioration of urban water quality. The pollution of the river by faecal material may be rather explained by several different sources, including open defecation practice, and inadequate wastewater and faecal sludge containment, collection and treatment (e.g., the presence of basic pit latrines located nearby the river and the studied wells). Further detailed analysis, comparing the seasonal fluctuation of faecal contamination with regional precipitation, is therefore recommended to confirm the suggested relationship between the contamination of surface and ground-water sources, and higher urban runoff and groundwater table, respectively. The investigated Kokolo Canal is of particular importance because it constitutes not only an important water supply for urban agriculture and domestic uses, but also for recreational activities of many children (Figure 3-3 B, C and D). After their day’s activities, many children are taking bath into this part of the contaminated Kokolo Canal, and are therefore exposed to high potential health hazards (see discussion in 3.2). The microbial contamination of rivers and wells into the urban environment of Kinshasa are overall increasing the potential risks of human infections either by direct uptake (drinking water), by possible source of bacterial 87 contamination in raw vegetables, or by contamination during recreational activities (Kilunga et al. 2016). According to the results of this study, it is strongly recommend to (1) urgently increase the proportion of population using safely managed sanitation services, (2) better protect the river banks, groundwater and shallow wells from faecal contamination, (3) monitor surface water and groundwater quality for faecal contaminants (and priority chemicals like trace metals), and to (4) increase the proportion of wastewater and faecal sludge generated by households (and economic activities) that is safely treated. Such objectives imply massive investment for improving the containment, collection and treatment/reuse of wastewater and faecal sludge. The SDG 6 indicators and monitoring framework (e.g., Targets 6.1, 6.2 and 6.3) can help to develop an adapted implementation strategy and to report on the access to safely managed sanitation and drinking water services; as well as to reduce the persistence and recurrence of epidemics in sub-rural communes of the city of Kinshasa. The method and approach developed in this study can provide a better understanding and assessment of the microbiological pollution of urban water resources in rapidly developing mega-cities of low and middle-income countries; under tropical conditions and in the lack of appropriate wastewater and faecal sludge treatment facilities, hygiene and sanitation systems. Such approach is therefore recommended in similar environment and in city planning process of poor urban and peri-urban communities, to further reduce the faecal contamination of natural sources of (surface and ground) waters that are used by urban dwellers without any treatment. 88 References Ahmed, W., Stewart, J., Gardner, T., Powell, D., Brooks, P., Sullivan, D., Tindale, N., 2007. “Sourcing faecal pollution: A combination of library-dependent and library- independent methods to identify human faecal pollution in non-sewered catchments.” Water Research 41, 3771-3779. Alm, E.W., Burke J., Spain, A. 2003. “Fecal indicator bacteria are abundant in wet sand and freshwater beaches.” Wat. Res. 37, 3978-3982. Bernhard, A.E., Field, K.G., 2000. “A PCR Assay To Discriminate Human and Ruminant Feces on the Basis of Host Differences in Bacteroides-Prevotella Genes Encoding 16S rRNA.” Applied and Environmental Microbiology 66, 4571-4574. Brinkmeyer, R., Amon, R.M.W., Schwarz, J.R., Saxton, T., et al. 2015. “Distribution and persistence of Escherichia coli and Enterococci in stream bed and bank sediments from two urban streams in Houston, TX.” Sc. Total Environ. 502, 650-658. Converse, R.R., Blackwood, A.D., Kirs, M., Griffith, J.F., Noble, R.T., 2009. “Rapid QPCR- based assay for fecal Bacteroides spp. as a tool for assessing fecal contamination in recreational waters.” Water Research 43, 4828-4837. Craig, D.L., Fallowfield, H.J., Cromar, N.J., 2004. “Use of microcosms to determine persistence of Escerichia coli in recreational coastal water and sediment and validation with in situ measurements.” J. App. Microbiol. 96, 922-930. Craig, D.L., Fallowfield, H.J., Cromar, N.J., 2002. “Enumeration of faecal coliforms from recreational coastal sites: evaluation of techniques for the separation of bacteria from sediments.” J. App. Microbiol. 93, 557-565. Davies, C.M., Long, J.A.H., Donald, M., Ashbolt, N., 1995. “Survival of fecal microorganisms in marine and freshwater sediments.” Appl Environm Microbiol 61, 1888-1896. Davis, K., Anderson, M.A., Yates, M.V., 2005. “Distribution of indicator bacteria in Canyon Lake, California.” Water Research 39, 1277-1288. EU, European Directive 2006/7/CE of the European Parliament and of the Council of 15 February 2006 concerning the management of bathing water quality and repealing Directive 76/160/EEC. 89 Haller, L., Poté, J., Loizeau, J.-L., Wildi, W., 2009. “Distribution and survival of faecal indicator bacteria in the sediments of the Bay of Vidy, Lake Geneva, Switzerland.” Ecological Indicators 9, 540-547. Kapembo, M. L., Laffite, A., Bokolo, M. K., Mbanga, A. L., Maya-Vangua, M. M., Otamonga, J.-P., Poté, J. 2016. “Evaluation of Water Quality from Suburban Shallow Wells Under Tropical Conditions According to the Seasonal Variation, Bumbu, Kinshasa, Democratic Republic of the Congo.” Exposure and Health, 8(4), 487-496. doi:10.1007/s12403-016-0213-y. Ke, D.B., F. Martineau, C., Menard, F.J., Picard, C., Ménard, P.H., Roy, M., 1999. Ouellette, and M.G. Bergeron, Development of a PCR assay for rapid detection of enterococci. Journal of clinical microbiology. 37, 3497-503. Kilunga, P. I., Kayembe, J. M., Laffite, A., Thevenon, F., et al., 2016. “The impact of hospital and urban wastewaters on the bacteriological contamination of the water resources in Kinshasa, Democratic Republic of Congo.” Journal of Environmental Science and Health, Part A, 51(12), 1034-1042. Laffite, A., Kilunga, P. I., Kayembe, J. M., Devarajan, N., Mulaji, C. K et al., 2016. “Hospital Effluents Are One of Several Sources of Metal, Antibiotic Resistance Genes, and Bacterial Markers Disseminated in Sub-Saharan Urban Rivers.” Frontiers in Microbiology, 7(1128). doi:10.3389/fmicb.2016.01128 Lee, C.M., Lin, T.Y., Lin, C., Kohbodi, G.A., Bhatt, A., Lee, R., Jay, J.A., 2006. “Persistence of fecal indicator bacteria in Santa Monica Bay beach sediments.” Wat. Res. 40, 2593- 2602. Martínez-Santos, P., Martín-Loeches, M., García-Castro, N., Solera, D., Díaz-Alcaide, S., Montero, E., García-Rincón, J., 2018. « A survey of domestic wells and pit latrines in rural settlements of Mali: Implications of on-site sanitation on the quality of water supplies”. Int. J. Hyg. Environm. Health 220, 1179-1189. Morrison, C., Bachoon, D., Gates, K., 2008. “Quantification of enterococci and Bifidobacteria in Georgia estuaries using conventional and molecular methods.” Water Research. 42, 4001-4009. Mubedi, J. I., Devarajan, N., Faucheur, S. L., et al., 2013. « Effects of untreated hospital effluents on the accumulation of toxic metals in sediments of receiving system under 90 tropical conditions: Case of South India and Democratic Republic of Congo.” Chemosphere, 93, 1070-1076. Mwanamoki, P. M., Devarajan, N., Niane, B., Ngelinkoto, P., et al., 2015. « Trace metal distributions in the sediments from river-reservoir systems: case of the Congo River and Lake Ma Vallée, Kinshasa (Democratic Republic of Congo).” Environmental Science and Pollution Research, 22, 586-597. Nienie, A.B., Periyasamy, S., Laffite, A., Ngelinkoto, P., et al., 2017, “Microbiological quality of water in a city with persistent and recurrent waterborne diseases under tropical sub- rural conditions: The case of Kikwit City, Democratic Republic of the Congo.” International Journal of Hygiene and Environmental Health 220, 820-828. Nienie, A.B., Sivalingamp., Laffite, A., Ngelinkoto, P., 2017a. «Seasonal variability of water quality by physicochemical indexes and traceable metals in suburban area in Kikwit, Democratic Republic of the Congo.” Int. Soil Wat. Conserv. Res. 5, 158-165 Noble, R.T., Leecaster, M.K., McGee, C.D., Weisberg, S.B., Ritter, K., 2004. “Comparison of bacterial indicator analysis methods in stormwater-affected coastal waters.” Water Research 38, 1183-1188. Nola, M., Nougang, M.E., Noah Ewoti O.V, Moungang, L,M., Krier, F., Chihib, N.E, 2013. “Detection of pathogenic Escherichia coli strains in groundwater in the Yaoundé region (Cameroon, Central Africa).” Water Environment Journal 27, 328-337. Poté, J., Haller, L., Kottelat, R., Sastre, V., Arpagaus, P, Wildi, W., 2009. “Persistence and growth of faecal culturable bacterial indicators in water column and sediments of Vidy Bay, Lake Geneva, Switzerland.” J Environ Sci 21, 62-69. Pritchard, M., Mkandawire, T., O’Neill, J.G., 2008. “Assessment of groundwater quality in shallow wells within the southern districts of Malawi.” Physics and Chemistry of the Earth Part B 33: 812-823. doi: 10.1016/j.pce.2008.06.036. Rochelle-Newall, E., Nguyen, T. M. H., Le, T. P. Q., Sengtaheuanghoung, O., Ribolzi, O., 2015. “A short review of fecal indicator bacteria in tropical aquatic ecosystems: knowledge gaps and future directions.” Frontiers in Microbiology, 6(308). doi:10.3389/fmicb.2015.00308. 91 Rodriguez-Alvareza, M.S., Weirb, M.H., Popee, J.M., Seghezzoc, L., Rajald, V.B., Salussoa, M.M., Morana, L.B., 2015. “Development of a relative risk model for drinking water regulation and design recommendations for a peri urban region of Argentina.” Int. J. Hyg. Environm. Health 218, 627-638. Sabat, G., Rose, P., Hickey, W.J., Harkin, J.M., 2000. “Selective and Sensitive Method for PCR Amplification of Escherichia coli 16S rRNA Genes in Soil.” Applied and Environmental Microbiology 66, 844-849. Scott, T.M., Jenkins, T.M., Lukasik, J., Rose, J.B., 2005. “Potential Use of a Host Associated Molecular Marker in Enterococcus faecium as an Index of Human Fecal Pollution.” Environmental Science & Technology 39, 283-287. Thevenon, F., Regier, N., Benagli, C., Tonolla, M., Adatte, T., Wildi, W., Poté, J., 2012. “Characterization of faecal indicator bacteria in sediments cores from the largest freshwater lake of Western Europe (Lake Geneva, Switzerland).” Ecotoxicology and Environmental Safety 78, 50-56. Tshibanda, J.B., Devarajan, N., Birane, N., Mwanamoki, P.M., Atibu, E.K., et al., 2014. “Microbiological and physicochemical characterization of water and sediment of an urban river: N’Djili River, Kinshasa, Democratic Republic of the Congo.” Sustainability 4-5, 47-54 UNEP., 2011. Post-Conflict Environmental Assessment in Democratic Republic of the Congo. Synthesis for policy makers. United Nations Environment Programme, Nairobi, Kenya. ISBN: 978-92-807-3226-9. US EPA., 2004. U.S. Environmental Protection Agency. (2004). http://www.epa.gov/owow/info/NewsNotes/pdf/73issue.pdf (accessed January 5, 2018). UN-Water., 2016. Integrated Monitoring Guide for SDG 6 Targets and global indicators. Version, 19 July 2016. WHO (World Health Organization) (2004) Guidelines for drinking-water quality, recommendations Geneva WHO (World Health Organization), 2011. Guidelines for drinking-water quality (4th edition, 541 pp). Geneva, Switzerland 92 WHO/UNICEF, 2017. Joint Monitoring Program for Water Supply, Sanitation and Hygiene (JMP) – 2017 Update and SDG Baselines. 93 CHAPTER 4 Hospital effluents are one of several sources of metal, antibiotic resistance genes and bacterial markers disseminated in Sub-Saharan urban rivers A similar version of this chapter was published under the following reference: Laffite, A., P. I. Kilunga, J. M. Kayembe, N. Devarajan, C. K. Mulaji, G. Giuliani, V. I. Slaveykova and J. Poté (2016). "Hospital Effluents Are One of Several Sources of Metal, Antibiotic Resistance Genes, and Bacterial Markers Disseminated in Sub-Saharan Urban Rivers." Frontiers in microbiology 7: 1128-1128. DOI: 10.3389/fmicb.2016.01128 95 Abstract Data concerning the occurrence of emerging biological contaminants such as antibiotic resistance genes (ARGs) and fecal indicator bacteria (FIB) in aquatic environments in Sub- Saharan African countries is limited. On the other hand, antibiotic resistance remains a worldwide problem which may pose serious potential risks to human and animal health. Consequently, there is a growing number of reports concerning the prevalence and dissemination of these contaminants into various environmental compartments. Sediments provide the opportunity to reconstruct the pollution history and evaluate impacts so this study investigates the abundance and distribution of toxic metals, FIB, and ARGs released from hospital effluent wastewaters and their presence in river sediments receiving systems. ARGs (blaTEM, blaCTX-M, blaSHV, and aadA), total bacterial load, and selected bacterial species FIB (E. coli, Enterococcus (ENT)) and Pseudomonas species (Psd) were quantified by targeting species specific genes using quantitative PCR (qPCR) in total DNA extracted from the sediments recovered from 4 hospital outlet pipes (HOP) and their river receiving systems in the City of Kinshasa in the Democratic Republic of the Congo. The results highlight the great concentration of toxic metals in HOP, reaching the values (in mg kg-1) of 47.9 (Cr), 213.6 (Cu), 1434.4 (Zn), 2.6 (Cd), 281.5 (Pb), and 13.6 (Hg). The results also highlight the highest (P˂0.05) values of 16S rRNA, FIB, and ARGs copy numbers in all sampling sites including upstream (control site), discharge point, and downstream of receiving rivers, indicating that the hospital effluent water is not an exclusive source of the biological contaminants entering the urban rivers. Significant correlation were observed between (i) all analyzed ARGs and total bacterial load (16S rRNA) 0.51 to 0.72 (p<0.001, n=65); (ii) ARGs (except blaTEM) and FIB and Psd 0.57 < R < 0.82 (p<0.001, n=65); and (iii) ARGs (except blaTEM) and toxic metals (Cd, Cr, Cu, and Zn) 0.44 to 0.72, (p<0.001, n=65). These findings demonstrate that several sources including hospital and urban wastewaters contribute to the spread of toxic metals and biological emerging contaminants in aquatic ecosystems. 96 4.1 Introduction Contamination of freshwater resources with anthropogenic pollutants is a growing concern of interest because safe and readily available water is needed for drinking, domestic use, food production, and recreational purposes (WHO 2015). Freshwater resource pollution by various contaminants including toxic metals, persistent organic pollutants, pathogenic organisms, antibiotic resistant bacteria (ARB), and antibiotic resistant genes (ARGs) is still a major problem in many parts of the world (Poté et al. 2008, Knapp et al. 2012, Brechet et al. 2014, Czekalski et al. 2014, Devarajan et al. 2015). The situation is particularly alarming in developing regions such as in Sub-Saharan Africa where most rivers, lakes, and lagoons are receiving untreated hospital and industrial effluent water, mining effluents, and urban storm water runoff affected by anthropogenic pollutants due to intensive and uncontrolled urbanization (Feng et al. 2004, Chatterjee et al. 2007, Gnandi et al. 2011, Atibu et al. 2013, Mwanamoki et al. 2014, Mwanamoki et al. 2015). Hospital effluents are a particular case of anthropogenic pollutants. Indeed, hospital wastewaters are complex mixtures of chemical and biological substances which are continually discharged (Barcelo and Barceló 2003, Boillot et al. 2008, Verlicchi et al. 2010). This mixture is the result of diagnostic laboratory and research activity waste and medicine excretion which include active principles from medicinal products and their metabolites, chemicals, disinfecting agents, specific detergents, radioactive markers, iodinated contrast media, nutrients, and bacteria and their antimicrobial resistance genes (Verlicchi et al. 2010). Particularly studied among these hospital contaminants are bacteria and their antimicrobial resistance genes because of their great ability to disseminate and their clinical and financial impact (Davies et al. 1995, Cosgrove 2006). The remarkable effectiveness of antibiotics has reduced mortality linked to bacterial diseases in only a few decades but their over- and mis-use has rapidly lead to a dramatic exponential increase in antibiotic resistance and multidrug-resistant bacteria throughout the world (Davies and Davies 2010, Fair and Tor 2014). The speed with which resistance has spread is explained by acquired resistance. In contrast to chromosomal resistance which is responsible for resistance to an antibiotic or an antibiotic class, acquired resistance by genetic material acquisition may be responsible for resistance against many antibiotics or antibiotic classes (INSERM 2013). This resistance is harbored by many human and animal, pathogenic, and potentially pathogenic bacteria and can easily be spread by either conjugation, 97 transformation, or transduction of resistance genes which are generally located on mobile elements (plasmids, transposons, integrons) (Carattoli 2009, Cambray et al. 2010, Aminov 2011). It is well known that both metals and antibiotic resistance genes are located on the same mobile elements, leading to a co-selection of ARGs by metals (Baker-Austin et al. 2006, Seiler and Berendonk 2012). Metal contaminations are widely spread in anthropogenic environment contributing on ARGs propagations (Sakan et al. 2009, Ji et al. 2012, Seiler and Berendonk 2012). The question of the environmental and human risks of an increasing release of bacteria carrying ARGs into the natural environment has been a subject of intense scientific and political debate in recent years. Consequently, there are a growing number of reports concerning the prevalence and dissemination of ARBs and ARGs into various environmental compartments (e.g. Kümmerer (2004), Martínez (2008), CDC (2013), WHO (2014), Devarajan et al. (2015). Effluents from hospitals, industry, municipal organizations, and urban/agricultural runoff in many developing countries represent a significant source of emerging contaminants (metals, ARGs, ARB) in the receiving environment as the effluents are discharged into sewer systems, rivers, lakes, and seas without prior treatment which may then accumulate in sediments (Spindler et al. 2012, Mwanamoki et al. 2014, Devarajan et al. 2015). Rivers and lakes are considered to be putative reservoirs of emerging contaminants (medicinal products, metals, ARGs) since they collect wastewaters containing various contaminants from various origins (Kümmerer 2004, Poté et al. 2008, Allen et al. 2010). Furthermore, the sediments may accumulate 100 to 1000 times as many heavy metals, FIB and ARGs as the overlying water (e.g. Poté et al., 2008; Haller et al., 2009; Thevenon et al. 2012a,b; Mubedi et al., 2013; Mwanamoki et al., 2014; 2015; Devarajan et al., 2015a,b) and offer the opportunity for reconstructing the pollution history and evaluating the impacts. Many studies have been performed to quantify ARB and ARGs in different environmental compartment around the word and explained the role of aquatic ecosystems as reservoirs of antibiotic resistance (e.g. Schwartz et al., 2003; Levy and Marshall, 2004; Kümmerer and Kummerer, 2004; Martínez, 2008; Stoll et al., 2012; Marti et al., 2014; WHO, 2014). Most of these studies established recommendations and hypotheses including the suggestion of further researches in different regions according to the source of drinking and recreational water, the practice of wastewater management, economic situation and sociocultural aspects of population, and climatic conditions. Nevertheless, many studies have not considered the influence of tropical conditions (e.g. in developing nations such as Sub- 98 Saharan African countries) on the accumulation of these emerging contaminants in aquatic environment, which can vary considerably with developed countries (under temperate conditions) (e.g. Czekalski et al., 2014; Devarajan et al., 2015). Consequently, little data is available on the assessment of heavy metals and neither is there much information to be found regarding quantitative and qualitative aspects of ARB as well as ARGs in the aquatic environment under tropical conditions which have average daily peak temperatures reaching 30°C. Measures to reduce the potential human and environmental risks caused by hazardous substances (such as toxic metals), ARB, and ARGs include their characterizations and selection of target ARB and ARG, identification of potential sources as well as risk assessment, the development of reliable surveillance and risk assessment procedures, and finally, the implementation of technological solutions that can prevent environmental contamination with ARB and ARGs (WHO 2014, Berendonk et al. 2015, Devarajan et al. 2015). The aim of the research presented in this paper is to assess the role of untreated hospital effluents discharged into freshwater receiving system under tropical conditions. This assessment was based on: (i) sediment physicochemical characterization including sediment grain size, total organic matter (OM; loss on ignition), and toxic metals including Cr, Co, Ni, Cu, Zn, As, Cd, Pb, and Hg - (ii) quantitative polymerase chain reaction (qPCR) on ARGs (blaTEM, blaCTX-M, blaSHV, and aadA), total bacterial load, and selected bacterial marker genes of fecal indicator bacteria (FIB) (E. coli and Enterococcus (ENT)) and Pseudomonas species (Psd). To the knowledge of the authors this is first report on the accumulation of emerging microbial contaminants in the sediments of freshwater receiving systems in a central African region and specifically in the city of Kinshasa, the capital of the Democratic Republic of the Congo. Nevertheless, it should be noticed that, one may argue that studies that analyze the DNA such as this study, while providing information on the presence/absence or even quantitative data but do not provide information on the expression of these ARGs (Lachmayr et al., 2009). However, the expression of genes is not the central query of this study when the purpose of this study is to address the evaluation of the tropical aquatic environment to serve as reservoirs of heavy metals and ARGs (that could be potentially transferred to other bacterial cells through horizontal gene transfer) (Devarajan et al., 2016). The parameters analyzed were correlated in order to identify the potential sources of receiving system contamination. 99 4.2 Material and Method 4.2.1 Study site and sampling Kinshasa, the capital and largest city in the Democratic Republic of the Congo (DRC) (Figure 4-1), is the 27th largest urban area in the world with 11,587,000 inhabitants and covering 9’965 km². The climate is classified as humid and dry with an average temperature of 21-30°C. The city has approximately 20 hospitals and various medical centers and polyclinics with different intrinsic characteristics (size, type of services, medical practices…). The wastewater effluents are discharged from the big hospitals into drainage systems and ejected into the urban river receiving systems without prior treatment. For small hospitals and many medical centers, wastewaters are directly rejected onto the soil, septic tanks, wells, or into rivers, and even using buckets. The selection of hospitals was based on their size relatively to the practice of wastewater management as described above, and their location near river receiving system. Three rivers receiving hospital effluent waters were selected for this study. They are affected by one very large hospital and three recent and innovative hospitals. No industry is located near the hospitals selected. Figure 4-1 Localization of the sampling site in the province of Kinshasa, Republic Democratic of Congo (Adapted from Google Maps) The sampling took place in December 2014. The surface sediments (0-4 cm layer) were collected from (i) outlet pipes (HOP) of the 4 hospitals selected, labelled H1, H2, H3, and H4. 100 The collection points were adjacent to the hospital effluent outlet pipe before discharge into rivers. This sampling point is named E, (ii) at the HOP discharge points into the rivers (point named RP), (iii) in the rivers 50 m upstream from HOP discharge points (point named US), and (iv) in the rivers 50 m downstream from HOP discharge points (point named DS). Approximately 400-500 g of sediment were taken from each site in triplicate. All samples were stored in an icebox at 4 ºC until shipping (Mubedi et al., 2013; Mwanamoki et al., 2015; Devarajan et al., 2015b) and analyzed within 2 weeks. 4.2.2 Sediment grain size, organic matter and water content The sediment particle grain size was measured using a Laser Coulter® LS-100 diffractometer (Beckman Coulter, Fullerton, CA, USA), following 5 min ultrasonic dispersal in deionized water according to the method described by Loizeau et al. (1994) . The sediment total organic matter (OM) content was estimated by loss on ignition at 550oC for 1h in a Salvis oven (Salvis AG, Emmenbrücke, Lucerne, Switzerland). Sediment total water content was measured by drying the samples at 60 °C for overnight and the weight loss was taken for the percentage of water content. 4.2.3 Toxic metal analysis Before being analyzed, sediment samples were lyophilized at -45°C after homogenization and air-drying at ambient room temperature. Toxic metals including Cr, Cu, Zn, Cd and Pb were determined by Inductively Coupled Plasma Mass Spectrometry (ICP-MS, Agilent model 7700 series) following the digestion of sediments in Teflon bombs heated to 150°C in analytical grade 2M HNO3 (Loizeau et al. 1994, Pardos et al. 2004, Poté et al. 2008). Multi-element standard solutions at different concentrations (0, 0.02, 1, 5, 20, 100, and 200 µg/L) were used for calibration. Total variation coefficients of triplicate sample measurements were under 5% and chemical blanks for the procedure were less than 2% of the sample signal. The metal concentrations of sediments were expressed in ppm (mg kg-1 dry weight sediment). Total Hg analysis was carried out using the Atomic Absorption Spectrophotometer (AAS) for mercury determination (Advanced Mercury Analyser; AMA 254, Altec s.r.l., Czech Rep.) following the method described by Hall and Pelchat (1997) and Ross-Barraclough et al. 101 (2002). The method is based on sample combustion, gold amalgamation, and AAS. The detection limit (3 SD blank) was 0.005 mg kg-1 and the reproducibility better than 2%. 4.2.4 Total DNA extraction Total DNA from sediment samples was extracted using the PowerSoil® DNA Isolation Kit (MoBio Laboratories, Carlsbad, USA) according to manufacturer’s instructions. DNA extraction was performed with three replicate sample (from the same sediment sample) to compensate for heterogeneity. The concentration of extracted DNA was measured using a Qubit Fluorimeter (Life Technologies Europe B.V., Zug, Switzerland). The isolated DNA was stored at -20 ˚C until used. 4.2.5 qPCR Quantification of selected genes in sediments: 16S rRNA, ARGs and FIB Quantification of ARGs (blaTEM, blaCTX-M, blaSHV and aadA), bacterial gene markers for E. coli, Enterococcus and Pseudomonas species, and 16S rRNA genes by qPCR was performed (qPCR reactions, control plasmids, calculation for absolute gene copy numbers (gene concentration) and the gene copy numbers normalized to 16S rRNA (abundance)) as previously described by Devarajan et al. (2015a). Briefly; genes were quantified with Eco qPCR system (Illumina, Switzerland) using KAPA SYBR® FAST qPCR Master Mix Universal Kit (KAPA Biosystems, USA). The primer sequences and reaction conditions are provided in Table 4-1. The following cycling parameters were applied: 10 min at 95 ˚C for the polymerase activation; followed by 40 cycles of 95 ˚C for 30 s, optimal Tm for 30 s and 72 ˚C for 30 s. The temperature melting curve profile was obtained using the following conditions; 95 ˚C for 30 s, optimal Tm for 30 s, followed by 95 ˚C for 30 s. Table 4-1 Primers used in this study Target organism/gene Primer Oligonucleotide sequence (5’-3’) Tm Size References (°C) (bp) Bacterial 16S rRNA 338F ACTCCTACGGGAGGCAGCAG 55 197 (Ovreas et al. 1997) 518R ATTACCGCGGCTGCTGG E.coli (uidA) Uida405F CAACGAACTGAACTGGCAGA 55 121 (Chern et al. 2011) Uida405R CATTACGCTGCGATGGAT ENT (16S rRNA) Ent376F GGACGMAAGTCTGACCGA 55 221 (Ram et al. 2004) Ent578R TTAAGAAACCGCCTGCGC Pseudomonas spp. Pse435F ACTTTAAGTTGGGAGGAAGGG 55 251 (Bergmark et al. 2012) Pse435R ACACAGGAAATTCCACCACCC blaTEM TEM-RT-F GCKGCCAACTTACTTCTGACAACG 55 247 (Sidrach Cardona et al. TEM-RT-R CTTTATCCGCCTCCATCCAGTCTA 2014) blaCTX-M blaCTX-M-rt-f ATTCCRGGCGAYCCGCGTGATACC 62 227 (Fujita et al. 2011) blaCTX-M-rt-r ACCGCGATATCGTTGGTGGTGCCAT blaSHV blaSHV-rt-f CGCTTTCCCATGATGAGCACCTTT 60 110 (Xi et al. 2009) 102 blaSHV-rt-r TCCTGCTGGCGATAGTGGATCTTT aadA aadA-F GCAGCGCAATGACATTCTTG 55 282 (Madsen et al. 2000) aadA-R ATCCTTCGGCGCGATTTTG All the reactions included negative (with no template DNA) and positive controls (10- fold serial dilutions of pGEM-T plasmid with respective target gene insert). All negative controls resulted either in no amplification or a threshold cycle (Ct) higher than the most diluted standard (pGEM-T plasmid). A sample was considered to be below the limit of detection (LOD) or negative for a target gene if ≥2 out of 3 technical replicates were negative or if sample Ct values were ≥Ct of negative controls. Samples above LOD were considered to be below the limit of quantification when the standard deviation of Ct values of methodological triplicates was 40.5 and their Ct value was higher than the Ct of the most diluted standard whose standard deviation of Ct values was ≤0.5. For each reaction the efficiency of the assay was measured using the slope of the standard curve measures (E = 10[-1/slope] – 1). The absolute copy number of each reaction was quantified by referring to the corresponding standard curve obtained by plotting the copy number of the constructed pGEM-T plasmid versus threshold cycles. The serial 10-fold dilutions of plasmid DNA containing the respective target gene copies were used for the standard curve. To emphasize the relative abundance of the resistance genes the concentrations of the gene copy numbers were presented as percentage of ‘‘copy number of a gene/copy number of 16S rRNA’’ for each sample. 4.2.6 Data analysis The 16S rRNA (total bacterial load), FIB and the selected marker genes and ARGs in the samples are expressed as “gene copy numbers” in per gram of dry sediment weight normalized to the DNA extraction yield. The “relative abundance” of the selected genetic marker genes (normalized to 16S rRNA) were emphasized by the ratio = (copy number of a gene) / (copy number of 16S rRNA) for each sample (Czekalski et al. 2014, Devarajan et al. 2015). A statistical treatment of data; Correlation matrix (Pearson), Principal component analysis (PCA) and its extensions to between (BGA) and within groups (WGA) analyses (ade4 package in R) was used to analyze the grouping hospital/sampling point samples by monitoring quality variables (metal content, FIB abundance, ARGs, mean grain size and OM content). The statistical software was R version 3.2.2 (R Core Team 2015). Linear fixed model were fitted to the data using the functionality of the package lme4 (Bates Douglas et al. 2015). Average 103 concentrations of gene copy numbers was modelled using the hospital and the sampling point as fixed effects and technical and biological effects as random effects. Significance of fixed effects was assessed by a t-test using a significance level of 5%. Model checking was based residual plots and normal probability checking using the raw residuals. Models were reduced using the likelihood ratio test. Pairwise comparison were evaluated based on adjusted p-values obtained using single-step method (Hothorn et al. 2008). 4.3 Results and Discussion 4.3.1 Sediment physicochemical parameters and metal content Sediment characteristics including particle grain-size and total organic matter (OM) are presented in Table 4-2. The sediment grain size and the OM varied substantially internally within the sampling sites (p˂0.05). Surface sediments of rivers in all sites studied are generally sandy-silt. The maximum value of clay observed at sites H1, H2, and H3 was less than 3%. H4 presented the maximum value of clay (11%). The same distribution was observed for sediment OM content (p˂0.05). The values ranged from 15.0-46.2% (H1), 4.7-9.4% (H2), and 0.7-8.7%. No great difference in OM was observed in the sampling sites of H4 (8.3-8.5%). Previous studies have reported that there are large variations in the distribution of sediment OM and grain size in freshwater, lakes, rivers, and reservoirs. The OM in non-contaminated freshwater sediments varies from 0.1-6.0% (Poté et al. 2008, Haller et al. 2009, Mubedi et al. 2013). The sediment from all sites (US: upstream; E: exit (outlet); RP: reject point (outlet discharge) and DS: downstream) of H1 and H4 are contaminated by organic matter. These results support the hypothesis that hospital effluents are one of many sources of contamination and that the contamination occurs at multiple points of entry along the river bank. 104 Table 4-2 Physico-chemical parameters of surface sediments from sampling points of each hospital site Sampling Mean Grain size Hospital Clay (%) Silt (%) Sand (%) OMc (%) pointb (µm) US 3,21 27,46 69,33 46,57 15,03 E 2,61 47,96 49,43 65,69 46,15 Hospital 1 RP 3,36 58,44 38,20 42,70 34,78 DS 1,31 25,45 73,24 139,10 18,10 US 0,41 14,00 85,59 157,20 6,05 E NDa ND ND ND ND Hospital 2 RP 0,48 13,46 86,06 229,10 9,37 DS 1,85 63,27 34,88 48,67 4,74 US 0,13 10,66 89,21 226,00 0,75 E 0,78 9,46 89,76 205,80 3,91 Hospital 3 RP 0,38 7,31 92,31 234,20 1,79 DS 0,27 27,47 72,26 135,20 8,72 US 5,68 38,71 55,61 52,67 8,33 E 0,37 58,23 41,40 58,71 8,40 Hospital 4 RP 10,83 57,51 31,66 24,04 8,44 DS 0,94 26,00 73,06 82,54 8,51 aND : not determinated bUS: upstream; E: exit; RP: reject point; DS: downstream cOM: organic matter The results of the toxic metal analysis are reported in Table 4-3. The maximum concentration was observed in sediments at H1, reaching values (in mg kg-1) of 47.9 (Cr), 213.6 (Cu), 1434.4 (Zn), 2.6 (Cd), 274.2 (Pb), and 13.6 (Hg). The sewages and rivers collect the majority of drained urban wastewater of the city of Kinshasa, are used as uncontrolled landfills for domestic solid wastes by the local population (Mavakala et al., 2016; Mwanamoki et al., 2015, Mubedi et al., 2013), explaining the presence of high metal concentrations in sediment. However, the presence of other non-identified sources (such as artisanal activities) and untreated hospital effluent water discharge cannot be excluded. In general, the concentration of toxic metals at points E, RP, and DS are higher for all sampling sites than the upstream sampling points indicating the effect of hospital effluent waters on the contamination of rivers by toxic metals 105 Table 4-3 Metal content of surface sediment samples from sampling point of each hospital site, analyzed by ICP-MS (expressed in mg. kg-1) Sampling Hospital Cr Co Ni Cu Zn Cd Pb Hg point US 34,32 4,29 14,24 107,45 884,61 1,79 189,62 1,43 E 45,54 4,55 17,42 213,59 1 004,47 1,88 137,08 13,60 Hospital 1 RP 47,87 4,97 17,42 204,13 1 077,17 2,07 124,40 3,94 DS 47,58 7,11 21,20 184,67 1 434,78 2,65 274,19 3,25 US 13,03 1,79 6,29 59,18 410,69 0,57 86,67 0,64 E ND ND ND ND ND ND ND ND Hospital 2 RP 11,96 1,81 13,48 34,18 304,92 0,54 281,52 0,44 DS 10,43 1,61 5,67 51,91 352,41 0,46 81,15 0,74 US 3,20 0,37 16,73 46,06 87,21 0,18 36,38 0,15 E 7,81 0,83 14,63 22,08 147,29 0,29 34,48 0,68 Hospital 3 RP 4,18 0,44 13,57 8,86 68,27 0,08 15,82 0,31 DS 4,97 0,62 3,79 22,65 153,77 0,22 40,75 0,37 US 15,36 0,99 4,85 9,38 71,99 0,10 14,96 0,39 E 27,92 2,68 9,98 47,54 365,75 0,52 75,98 0,56 Hospital 4 RP 21,91 1,20 5,96 12,66 98,75 0,16 20,98 0,51 DS 15,23 1,13 5,10 12,86 100,30 0,14 20,73 0,44 SQGsa 37,30 35,70 123,00 0,60 35,00 0,17 b PEL 90,00 197,00 315,00 3,50 91,30 0,49 aSediment quality guidelines (mg.kg-1) bProbable effect level (mg.kg-1) In bold: values above SQGs Cr: chromium; Co: cobalt; Ni: nickel; Cu: copper; Zn: zinc; Cd: cadmium; Pb: lead; Hg: mercury . Metal accumulated in sediments is one of a good indicators for predicting the deterioration of a contaminated environment by inorganic pollutants. The release of heavy metals into the aquatic ecosystem can lead to the pollution of water resources and may place aquatic organisms and human health at risk. The main human and environmental risk is remobilization of the contaminants and their return to the hydrosphere either by sediment re- suspension or by infiltration into groundwater (Wildi et al., 2004; Poté et al., 2008). The evaluation of the potential deleterious effects of the metals towards benthic fauna, which is based on sediment quality guidelines (SQG) (CCME 1999, MacDonald et al. 2000, Long et al. 2006) provides an estimate of the hazard that the sediments may represent to the local biota. The authors proposed a “Threshold effect concentration” (TEC) for specific metals which is the 106 level above which an organism is affected (or responds) and below which it does not, and a “probable effect level” (PEL), a contaminant level that is likely to have an adverse effect on biota (Mwanamoki et al., 2014). According to the SQGs and PEL (Table 4-3), the concentration of Cr, Cu, Zn, Pb, and Hg in sediment may present a potential toxic effect on indigenous fauna and flora living in these aquatic environments. Furthermore, the high PEL values observed upstream of hospital effluent discharges indicate that hospital effluents are only one of several sources of contamination. Indeed, many anthropogenic activities (i.e. industrial, sewage, agricultural land field, mining…) are known to be responsible of heavy metal release in the environment (Poté et al. 2008, Lim et al. 2013, Devarajan et al. 2015, Niane et al. 2015) and may be responsible of the high metal values observed upstream HOP effluents discharge. 4.3.2 Abundance of bacterial population The total bacterial load in sediment samples is presented in Figure 4-2 (based on 16S rRNA gene copy numbers). In general, 16S rRNA gene copy numbers varied significantly between sampling sites (p< 0.05) with values of log copy numbers g−1 of dry sediment). The 16S rRNA gene copy numbers in samples from H1 were several orders of magnitude higher than those observed in others hospitals. However, these results are in conformity with those found in sediment from contaminated sediments receiving WWTP effluent waters (Devarajan et al. 2015). Furthermore, the total bacterial load after the effluent discharge (RP) for H1 and H4 were respectively 10 and 7.85 times higher than the values observed in their respective control sites (US). In contrast, the total bacterial load in the points RP for H2 and H3 was 3.7 to 6.8 times lower than their respective control site (US), indicating the possible presence of other sources of contamination. 107 Figure 4-2 Raw 16S rRNA copy number detected in hospital receiving systems (16S rRNA gene copy number / g of DS) at each sampling point. Significant stars represent significant variation between US and RP point, ***-p<0.01, ** 0.01>p>0.01 and * 0.1>p>0.05. For the pairs not marked the statistical difference between US and RP was statistically insignificant. The line in each box marks the media and boxes: 25th and 75th percentiles; whiskers: 5th and 95th percentiles and outliers± 1.5 *IQ. US: upstream; E: hospital outlet effluent; RP: reject point; DS: downstream The average copy number of FIB bacterial marker genes including E. coli, ENT, and Psd in the sediment samples is presented in Figure 4-3. The bacterial density varies considerably depending on the sampling point and the type of hospital. For example, the E. coli range at the US site (log copy numbers g−1 of dry sediment) was 6.44, 6.30, 6.67, and 4.21 for H1, H2, H3, and H4 respectively; ENT range (log copy numbers g−1 of dry sediment) of 7.28, 7.59, 7.87, and 5.41 for H1, H2, H3, and H4 respectively; and Psd range (log copy numbers g−1 of dry sediment) was 6.96, 6.94, 7.10, and 5.49 for H1, H2, H3, and H4 respectively. The input of hospital wastewater in the receiving system lead to a 6, 1, and 288 times greater abundance of E. coli for H1, H2, and H4 respectively. In H3 RP sampling point, a 6.2 times decrease in E. coli abundance was measured after wastewater discharge showing that H3 wastewater did not contribute more than the environmental enrichment in E. coli. 108 Figure 4-3 Raw FIB copy number detected in hospital receiving systems at each sampling point. Significant stars represent significant variation between US and RP point, ***-p<0.01, ** 0.01>p>0.01 and * 0.1>p>0.05. For the pairs not marked the statistical difference between US and RP was statistically insignificant. The line in each box marks the media and boxes: 25th and 75th percentiles; whiskers: 5th and 95th percentiles and outliers ± 1.5*IQR. US: upstream; E: hospital outlet effluent; RP: reject point; DS: downstream ENT: Enterococcus; Psd: Pseudomonas species Raw load of selected bacterial marker genes followed the 16S rRNA trend. To avoid inconstancies between qPCR assays, including suboptimal efficiencies, selected bacterial species marker genes and ARGs were normalized using 16S rRNA (Figure 4-4and Figure 4-5). The greatest abundance of bacterial populations was recorded at H1 and H4 sampling sites (except E. coli for H1 and Psd for H4). H2 and H3 did not show any significant increase in bacterial marker genes abundance, which was also the case for 16S rRNA abundance (0.20 109 contribute more to already environmental abundance and the observed decrease in FIB abundance may be due to contaminant dilution and transport downstream by river flow (Knapp et al. 2012, Chen et al. 2013). Figure 4-4 Normalized FIB copy number Figure 4-5 Normalized ARGs copy number detected in hospital receiving systems at detected in hospital receiving systems at each sampling point. each sampling point. *Significant stars represent significant variation between US and RP point, ***-p<0.01, ** 0.01>p>0.01 and * 0.1>p>0.05. For the pairs not marked the statistical difference between US and RP was statistically insignificant. The line in each box marks the media and boxes: 25th and 75th percentiles; whiskers: 5th and 95th percentiles and outliers ± 1.5*IQR. US: upstream; E: hospital outlet effluent; RP: reject point; DS: downstream FIB including E. coli and ENT are commonly used to assess microbial safety of aquatic systems. It is well known that many E.coli and ENT are also responsible for numerous health care-associated infections of the bloodstream, urinary tract, and surgical incision sites (Alm et al. 2014). To develop this high level of pathogenicity, these bacteria have acquired islands of pathogenicity including antibiotic, heavy metal resistance genes, and virulence factors. Several studies in other parts of the world have revealed the presence of pathogenic micro-organisms which are multi-resistant to antibiotics in hospital effluents (Emmanuel et al. 2009). In the 110 absence of wastewater treatment as is the case in the hospitals in the study presented in this paper, these bacteria will be discharged directly into aquatic receiving systems. It is well known that FIB are able to survive and proliferate in sediments as sediments provide favorable conditions for proliferation and growth (Poté et al. 2009). Furthermore, FIB have a great ability to acquire ARG (Levy and Marshall 2004) and are able to transfer their resistance to autochthonous bacteria by HGT (Sidrach Cardona et al. 2014). Consequently, the discharge of raw hospital wastewater could lead to an environmental reservoir of clinical resistant bacteria and their associated genes developing in the sediment (Marti et al. 2014). 4.3.3 Quantification of antibiotic resistance genes qPCR was performed to quantify the selected ARGs conferring resistance to ß-lactam (blaTEM, blaCTX-M, and blaSHV) and aminoglycoside (aadA) in DNA extracted from the sediment samples. ARG selection was based on various criteria including (Devarajan et al., 2016): (i) clinically relevant genes (human risk); (ii) genes conferring resistance to frequently used antibiotics; (iii) ARGs previously reported in mobile genetic elements; and (iv) the antibiotics used in 6 pilot hospitals in Kinshasa (Nzolo et al. 2013). The raw gene copy number (ARGs g- 1 of dry sediment) was used to estimate the general changes in ARGs level in receiving systems. The results are presented in Figure 4-6. The ARGs copy number (ARGs g-1 of dry sediment) for aadA, blaCTX-M, blaSHV, and blaTEM varied respectively from 4.64 to 7.83, 4.67 to 5.01, 3.92 to 4.66, and 4.23 to 4.86 at US sites, and from 5.33 to 9.24, 4.55 to 5.61, 3.76 to 6.17, and 4.36 to 5.46 at RP/DS sampling sites. The great abundance of blaTEM and aadA genes at all sampling sites (US, E, RP, and DS) could be explained by their ubiquitous presence as housekeeping genes, which has previously been shown to occur frequently among soil bacteria as well as by their presence in sewage and effluent receiving systems (Demanèche et al. 2008, Thevenon et al. 2012, Suzuki et al. 2015). A relevant increase in ARG level after wastewater discharge is only observed in H1 sediments (p<0.05) with a 25.6, 45.0, and 148 times increase observed between US and RP sampling points for blaCTX-M, blaSHV, and aadA respectively. For the others hospitals, the general trend of ARGs increased after wastewater discharge was no longer observed. The abundance of ARGs and associated bacteria in a receiving system may vary depending on the size of the hospital and type of services. 111 Figure 4-6 Raw ARGs copy number detected in hospital receiving systems at each sampling point. Significant stars represent significant variation between US and RP point, ***-p<0.01, ** 0.01>p>0.01 and * 0.1>p>0.05. For the pairs not marked the statistical difference between US and RP was statistically insignificant. The line in each box marks the media and boxes: 25th and 75th percentiles; whiskers: 5th and 95th percentiles and outliers ± 1.5*IQR. US: upstream; E: hospital outlet effluent; RP: reject point; DS: downstream The normalized/relative abundances of 16S rRNA were found in order to quantify relative change in ARGs abundances, that is, whether or not more or fewer ARGs appear per microbial genome (Laht et al. 2014). The relative abundance of ARGs in the sediment samples is presented in Figure 4-5. In general, the influence of hospital effluents was not observed in the relative abundance of ARGs copy numbers in receiving systems. Some specific cases such as aadA in H4, blaSHV in H1, and blaCTX-M in H2 increased significantly after hospital effluent discharge (p<0.01). These data suggest that increases in ARGs abundances are directly related to wastewater discharge depending on hospital type and disposal practices, although unknown sources and/or causes also exist (Graham et al. 2011, Marti et al. 2014). The prevalence of microbial contaminants in the control sites (upstream) could be explained by, for example, input from major activities such as agricultural runoff, open defecation, urban discharge, and other anthropogenic activities along the river banks, which receives considerable amounts of wastewaters (Dekov et al. 1998, Tshibanda et al. 2014). The abundance of total bacterial load, 112 ARGs, and FIB showed a relevant increase after wastewater discharge but the results highlight that hospital wastewater effluents are not the only source of micropollutant accumulation. Interestingly, the relative abundances of ARGs observed in this study were greater than in other studies performed in a similar environment under tropical conditions (Graham et al. 2011, Devarajan et al. 2015) and can be compared to data obtained in industrialized countries (Devarajan et al. 2015). However, the lack of background information and the knowledge gap in our pristine study site does not help in understanding the trends observed. ESBLs are mostly TEM, SHV, and CTX-M derivatives. The great abundance of clinically relevant ARGs such as blaCTX-M and blaSHV in the sediments studied may indicate the possible presence of ESBL in these systems (Poirel et al. 2012). The presence of ESBL in rivers (such as the sites studied) is highly alarming because the majority of ESBL are resistant to first line antibiotics but are also resistant to a large number of relevant antibiotics (Tacao et al. 2014). Furthermore, such a large abundance of ARGs represents a serious threat from resistance propagation because gene exchange can take place in sediment between both dead and living bacteria (Mao et al. 2014). It has been determined that approximately 90 bacterial species have natural transformability competences. Among them are many human pathogens, including the genera Campylobacter, Haemophilus, Helicobacter, Nesseiria, Pseudomomas, Staphylococcus, and Streptococcus (Mao et al. 2014). So the reservoir of ARGs in river sediments can be easily used by other bacteria to become ever more resistant. The abundance of ARGs in the receiving systems reported in this study can be considered as alarming. The results indicate clearly that sediment receiving system under tropical condition can act as reservoirs of ARGs including blaTEM, blaCTX-M, blaSHV, and aadA. It has been demonstrated that the intensive use of antibiotics for humans, animals, and agricultural purpose has led to the release (e.g. through the disposal of human and animal wastes) of ARB and ARGs into soil and aquatic environment (Sommer et al., 2009; Martínez, 2008). In the study region as well as in many Sub-Saharan African countries, there is no regulation for the use of antibiotics in humans, animals as well as for agricultural purpose, and data concerning the occurrence of ARGs, ARBs and FIB in aquatic environments is limited (WHO, 2014; Devarajan et al., 2016). Additionally, there is no policies and management tools to facilitate the urban wastewater treatment in study region. Thus, according to the results of this study, we strongly recommend the prudence and regulation for the use of antibiotics in human and animal consumption, to limit spread of ARGs and ARB into the environment. Furthermore, the need of a strategy for hospital and urban wastewater treatment. 113 4.3.4 Statistical correlation Correlation analysis between total bacterial load, FIB, ARGs, toxic metals, total organic matter, and sediment grain size was carried out to determine potential links between both parameters analyzed and possible origins of contaminants with the results being presented in Table 4-4. Total organic matter content, metal concentrations, bacterial indicator genetic markers (except E.coli), and ARGs were mostly significantly, positively, and mutually correlated. Nevertheless, all ARGs studied at hospital and sampling points are significantly correlated with total bacterial load (16S rRNA): 0.51 to 0.72 (p<0.001, n=65). ARGs (except for blaTEM) have a positive correlation with E. coli, ENT, and Psd (0.57 < R < 0.82, p<0.05, n=65) and the metals (Cd, Cr, Cu, Hg, and Zn): (0.37 < R < 0.71, p<0.001, n=65). Strong positive and mutual correlation was observed between 16S rRNA, E. coli, ENT, and Psd with ARGs (except blaTEM). These results indicate that these biological contaminants could originate from common sources and they are carried to the receiving system by common transporters (Thevenon et al. 2012, Devarajan et al. 2015). In addition, there was a positive correlation between total organic content and the metals in sediments. This observation is also supported by the fact that the contaminants are attached to both large organic and small inorganic particles such as clay and they could behave in a similar way in transporting contaminants to the receiving system (Poté et al. 2008, Zhao et al. 2015). Table 4-4 Pearson correlation Cd Cr Cu Grain size Hg 16S rRNA aadA CTX-M E.coli ENT Psd SHV TEM OM Pb Zn Cd 0.90 0.95 -0.24 0.64 0.59 0.66 0.54 -0.10 0.47 0.64 0.60 0.25 0.76 0.63 0.99 Cr 0.89 -0.52 0.66 0.61 0.59 0.44 -0.05 0.51 0.67 0.67 0.35 0.83 0.43 0.89 Cu -0.29 0.78 0.67 0.71 0.57 -0.11 0.54 0.78 0.72 0.25 0.87 0.49 0.95 Grain size -0.26 -0.33 -0.01 0.03 -0.05 -0.20 -0.29 -0.32 -0.33 -0.41 0.23 -0.26 Hg 0.51 0.52 0.37 -0.05 0.46 0.70 0.63 0.08 0.89 0.25 0.63 16S rRNA 0.72 0.67 -0.19 0.83 0.80 0.63 0.51 0.72 0.24 0.57 aadA 0.58 -0.36 0.74 0.73 0.66 0.18 0.63 0.43 0.65 CTX-M -0.11 0.57 0.62 0.56 0.24 0.59 0.52 0.51 E.coli -0.34 -0.17 -0.13 -0.04 -0.04 -0.12 -0.11 ENT 0.73 0.59 0.47 0.59 0.27 0.45 Psd 0.82 0.26 0.86 0.21 0.61 SHV 0.15 0.79 0.19 0.57 TEM 0.23 0.12 0.25 OM 0.32 0.74 Pb 0.63 Zn BGA on PCA analysis showed that sampling points varied greatly between hospital sites revealing the impact of (i) initial background level in ARGs, heavy metals, and FIB and (ii) the hospital type (Figure 4-7-left). To analyze effluent effect independently of the hospital, sampling points were decomposed before fitting (WGA on PCA using hospital as ‘within group’ factor) (Figure 4-7-right). The results show a large increase in ARGs, FIB, toxic metals, 114 and OM for hospital H1 and H4 linked to wastewater discharge. Lastly, these results indicate that river receiving systems could depend on hospital practices but also that river sediments are already significantly contaminated by unknown sources. As the rivers flow through the City of Kinshasa, additional pollutant sources such as domestic sewage, uncontrolled landfill and artisanal activities located in the banks of rivers can probably explain the presence of contaminants accumulation in sediment (Mubedi et al. 2013, Ngelinkoto et al. 2014, Mwanamoki et al. 2015). Figure 4-7 Grouping of each sampling point according to ARGs, FIB, toxic metals, grain size and OM. A – plot using between group analyses to discriminate each points in various hospital receiving system. B – same samples plotted after decomposing differences in each sampling point by within group analysis. Right upper panels show correlation with variables (PCA variable scores). 4.4 Conclusion The research presented in this paper investigates the abundance and dissemination of metal, FIB and ARGs released from hospital effluents into the urban river receiving systems. It’s important to note that one of the main concerns consisting on the evaluation of the degree to which river receiving system under tropical conditions can act as reservoir of FIB and ARGs. Results demonstrate accumulation of toxic metals, E. coli, Enterococcus, and Pseudomonas species as well as ARGs, indicating that river receiving systems under tropical conditions 115 (developing countries such as our study region) can act as a reservoir for metals and emerging microbial contaminants such as FIB and ARGs which can be transferred to human pathogens. Thus, the river receiving systems under tropical conditions which has average daily peak temperatures reaching 30°C, could potentially favor the transfer of mobile genetic elements carrying ARGs to susceptible bacterial pathogens. On the other hand, the presence of higher values of FIB and ARGs in sediment samples located upstream of the hospital outlet discharges (control sites) indicates that the hospital effluent wastewaters are not the only source of deterioration of bacteriological quality of studied rivers. The pollution in the cases studied in this paper may be explained by probable multiple diffuse pollution sources including open defecation, uncontrolled landfills, unregulated effluent discharges, and inadequate sewage collection near the sites studied. Rivers in most developing nations (such as our study region) serve as a basic network for human and animal consumption as well as irrigation for fresh urban produces. High values of metals, FIB and ARGs observed in river receiving systems indicate the human and environmental potential risks. However, further studies are required to find the pathways used by ARGs to spread, exploring their potential transfer into clinically relevant bacteria and human commensal as well as assessing the human exposure and environment potential risks. To our knowledge, this is the first study to be performed in the region regarding the quantification of ARGs in receiving systems. The quantification of FIB and ARGs as performed in this study can therefore facilitate improved risk assessments for the prudent use of antibiotics in human, animal and agriculture, and provide baseline information for developing strategies (such as hospital and urban wastewater treatment) to limit the spread of these emerging contaminants under tropical conditions. 116 References Allen, H. K., J. Donato, H. H. Wang, K. A. Cloud Hansen, J. Davies and J. Handelsman (2010). "Call of the wild: antibiotic resistance genes in natural environments." Nature Reviews Microbiology 8(4): 251-259. Alm, E. W., D. Zimbler, E. Callahan and E. Plomaritis (2014). "Patterns and persistence of antibiotic resistance in faecal indicator bacteria from freshwater recreational beaches." Journal of Applied Microbiology 117(1): 273-285. Aminov, R. I. R. (2011). "Horizontal gene exchange in environmental microbiota." Frontiers in microbiology 2: 158. Atibu, E. K., N. Devarajan, F. Thevenon, P. M. Mwanamoki, J. B. Tshibanda, P. T. Mpiana, K. Prabakar, J. I. Mubedi, W. Wildi and J. Poté (2013). "Concentration of metals in surface water and sediment of Luilu and Musonoie Rivers, Kolwezi-Katanga, Democratic Republic of Congo." Applied Geochemistry 39: 26-32. Baker-Austin, C., M. S. Wright, R. Stepanauskas and J. V. McArthur (2006). "Co-selection of antibiotic and metal resistance." Trends in Microbiology 14(4): 176-182. Barcelo, D. and D. Barceló (2003). "Emerging pollutants in water analysis." TrAC. Trends in analytical chemistry 22(10): xiv-xvi. Bates Douglas, Maechler Martin, W. S. Bolker Ben, Christensen Rune Haubo Bojesen, Singmann Henrik, Dai Bin and GrothendieckGabor (2015). Linear Mixed-Effects Models using 'Eigen' and S4. Berendonk, T. U., C. M. Manaia, C. Merlin, D. Fatta-Kassinos, E. Cytryn, F. Walsh, H. Bürgmann, H. Sørum, M. Norström, M.-N. Pons, N. Kreuzinger, P. Huovinen, S. Stefani, T. Schwartz, V. Kisand, F. Baquero and J. L. Martinez (2015). "Tackling antibiotic resistance: the environmental framework." Nat Rev Microbiol 13(5): 310- 317. Bergmark, L. L., P. H. P. H. B. Poulsen, W. A. W. A. Al-Soud, A. A. Norman, L. H. L. Hansen, W. A. W. Al Soud and S. J. S. Sørensen (2012). "Assessment of the specificity of Burkholderia and Pseudomonas qPCR assays for detection of these genera in soil using 454 pyrosequencing." FEMS microbiology letters 333(1): 77-84. Boillot, C., C. Bazin, F. Tissot-Guerraz, J. Droguet, M. Perraud, J. C. Cetre, D. Trepo and Y. Perrodin (2008). "Daily physicochemical, microbiological and ecotoxicological 117 fluctuations of a hospital effluent according to technical and care activities." Sci Total Environ 403(1-3): 113-129. Brechet, C., J. Plantin, M. Sauget, M. Thouverez, D. Talon, P. Cholley, C. Guyeux, D. Hocquet and X. Bertrand (2014). "Wastewater Treatment Plants Release Large Amounts of Extended-Spectrum beta-Lactamase-Producing Escherichia coli Into the Environment." Clinical Infectious Diseases 58(12): 1658-1665. Cambray, G. G., A. A.-M. Guerout and D. D. Mazel (2010). "Integrons." Annual review of genetics 44: 141-166. Carattoli, A. (2009). "Resistance Plasmid Families in Enterobacteriaceae." Antimicrobial Agents and Chemotherapy 53(6): 2227-2238. CCME (1999). Recommendation canadiennes pour la qualité des sédiments. C. C. o. M. o. t. Environment. CDC (2013). ANTIBIOTIC RESISTANCE THREATS in the United States, 2013. C. f. D. C. a. Prevention. Chatterjee, M., E. V. Silva Filho, S. K. Sarkar, S. M. Sella, A. Bhattacharya, K. K. Satpathy, M. V. R. Prasad, S. Chakraborty and B. D. Bhattacharya (2007). "Distribution and possible source of trace elements in the sediment cores of a tropical macrotidal estuary and their ecotoxicological significance." Environment International 33(3): 346-356. Chen, B., X. Liang, X. Huang, T. Zhang and X. Li (2013). "Differentiating anthropogenic impacts on ARGs in the Pearl River Estuary by using suitable gene indicators." Water Research 47(8): 2811-2820. Chern, E. C., S. Siefring, J. Paar, M. Doolittle and R. A. Haugland (2011). "Comparison of quantitative PCR assays for Escherichia coli targeting ribosomal RNA and single copy genes." Letters in applied microbiology 52(3): 298-306. Cosgrove, S. E. S. (2006). "The relationship between antimicrobial resistance and patient outcomes: mortality, length of hospital stay, and health care costs." Clinical infectious diseases 42 Suppl 2: S82-89. Czekalski, N., E. Gascon Diez and H. Burgmann (2014). "Wastewater as a point source of antibiotic-resistance genes in the sediment of a freshwater lake." ISME J 8(7): 1381- 1390. Davies, C. M., J. A. H. Long, M. Donald and N. J. Ashbolt (1995). "Survival of Fecal Microorganisms in Marine and Fresh-Water Sediments." Applied and Environmental Microbiology 61(5): 1888-1896. 118 Davies, J. J. and D. D. Davies (2010). "Origins and evolution of antibiotic resistance." Microbiology and molecular biology reviews 74(3): 417-433. Dekov, V. M., F. Araujo, R. Van Grieken, V. Subramanian and R. Vangrieken (1998). "Chemical composition of sediments and suspended matter from the Cauvery and Brahmaputra rivers (India)." Science of the total environment 212(2-3): 89-105. Demanèche, S., H. Sanguin, J. Poté, E. Navarro, D. Bernillon, P. Mavingui, W. Wildi, T. M. Vogel and P. Simonet (2008). "Antibiotic-resistant soil bacteria in transgenic plant fields." Proc Natl Acad Sci U S A 105(10): 3957-3962. Devarajan, N., A. Laffite, N. D. Graham, M. Meijer, K. Prabakar, J. I. Mubedi, V. Elongo, P. T. Mpiana, B. W. Ibelings, W. Wildi and J. Pote (2015). "Accumulation of clinically relevant antibiotic-resistance genes, bacterial load, and metals in freshwater lake sediments in Central Europe." Environ Sci Technol 49(11): 6528-6537. Devarajan, N., A. Laffite, P. Ngelikoto, V. Elongo, K. Prabakar, J. Mubedi, P. M. Piana, W. Wildi and J. Poté (2015). "Hospital and urban effluent waters as a source of accumulation of toxic metals in the sediment receiving system of the Cauvery River, Tiruchirappalli, Tamil Nadu, India." Environmental Science and Pollution Research 22(17): 12941-12950. E. M. Hall, G. and P. Pelchat (1997). "Evaluation of a Direct Solid Sampling Atomic Absorption Spectrometer for the Trace Determination of Mercury in Geological Samples." Analyst 122(9): 921-924. Emmanuel, E. E., Y. Y. Perrodin and M. G. M. Pierre (2009). "Groundwater contamination by microbiological and chemical substances released from hospital wastewater: Health risk assessment for drinking water consumers." Environment international 35(4): 718- 726. Fair, R. J. and Y. Tor (2014). "Antibiotics and bacterial resistance in the 21st century." Perspect Medicin Chem 6: 25-64. Feng, H., X. Han, W. Zhang and L. Yu (2004). "A preliminary study of heavy metal contamination in Yangtze River intertidal zone due to urbanization." Marine Pollution Bulletin 49(11–12): 910-915. Fujita, S., K. Yosizaki, T. Ogushi, K. Uechi, Y. Takemori and Y. Senda (2011). "Rapid Identification of Gram-Negative Bacteria with and without CTX-M Extended- Spectrum -Lactamase from Positive Blood Culture Bottles by PCR Followed by Microchip Gel Electrophoresis." Journal of clinical microbiology 49(4): 1483-1488. 119 Gnandi, K., S. Han, M. H. Rezaie-Boroon, M. Porrachia and D. Deheyn (2011). "Increased Bioavailability of Mercury in the Lagoons of Lomé, Togo: The Possible Role of Dredging." AMBIO 40(1): 26-42. Graham, D. W., S. Olivares-Rieumont, C. W. Knapp, L. Lima, D. Werner and E. Bowen (2011). "Antibiotic Resistance Gene Abundances Associated with Waste Discharges to the Almendares River near Havana, Cuba." Environmental Science & Technology 45(2): 418-424. Graham, J. P. and M. L. Polizzotto (2013). "Pit Latrines and Their Impacts on Groundwater Quality: A Systematic Review." Environmental health perspectives 121(5): 521-530. Haller, L. L., J. J. Pote, J. J.-L. Loizeau and W. W. Wildi (2009). "Distribution and survival of faecal indicator bacteria in the sediments of the Bay of Vidy, Lake Geneva, Switzerland." Ecological indicators 9(3): 540-547. Hothorn, T. T., F. F. Bretz and P. P. Westfall (2008). "Simultaneous Inference in General Parametric Models." Biometrical journal 50(3): 346-363. INSERM. (2013). "Résistance aux antibiotiques." Retrieved 10 octobre, 2015, from http://www.inserm.fr/thematiques/immunologie-inflammation-infectiologie-et- microbiologie/dossiers-d-information/resistance-aux-antibiotiques. Ji, X., Q. Shen, F. Liu, J. Ma, G. Xu, Y. Wang and M. Wu (2012). "Antibiotic resistance gene abundances associated with antibiotics and heavy metals in animal manures and agricultural soils adjacent to feedlots in Shanghai; China." Journal of hazardous materials 235-236: 178-185. Knapp, C. W., L. Lima, S. Olivares-Rieumont, E. Bowen, D. Werner and D. W. Graham (2012). "Seasonal variations in antibiotic resistance gene transport in the Almendares River, Havana, Cuba." Frontiers in Microbiology 3. Kümmerer, K. and K. Kummerer (2004). "Resistance in the environment." Journal of antimicrobial chemotherapy 54(2): 311-320. Laht, M., A. Karkman, V. Voolaid, C. Ritz, T. Tenson, M. Virta and V. Kisand (2014). "Abundances of Tetracycline, Sulphonamide and Beta-Lactam Antibiotic Resistance Genes in Conventional Wastewater Treatment Plants (WWTPs) with Different Waste Load." PLoS ONE 9(8): e103705. Levy, S. B. S. and B. B. Marshall (2004). "Antibacterial resistance worldwide: causes, challenges and responses." Nature Medicine 10(12 Suppl): S122-S129. Lim, W. Y., A. Z. Aris and H. T. H. Tengku (2013) "Spatial Geochemical Distribution and Sources of Heavy Metals in the Sediment of Langat River, Western Peninsular 120 Malaysia." Environmental forensics 14, 133-145 DOI: 10.1080/15275922.2013.781078. Loizeau, J. L., D. Arbouille, S. Santiago and J. P. Vernet (1994). "Evaluation of a wide range laser diffraction grain size analyser for use with sediments." Sedimentology 41(2): 353-361. Long, E. R., C. G. Ingersoll and D. D. MacDonald (2006) "Calculation and uses of mean sediment quality guideline quotients: a critical review." Environ Sci Technol 40, 1726-1736 DOI: 10.1021/es058012d. MacDonald, D. D., C. G. Ingersoll and T. A. Berger (2000). "Development and evaluation of consensus-based sediment quality guidelines for freshwater ecosystems." Archives of environmental contamination and toxicology 39(1): 20-31. Madsen, L. L., F. M. F. Aarestrup and J. E. J. Olsen (2000). "Characterisation of streptomycin resistance determinants in Danish isolates of Salmonella Typhimurium." Veterinary microbiology 75(1): 73-82. Mao, D. D., Y. D. Luo, J. J. Mathieu, Q. Q. Wang, L. L. Feng, Q. Q. Mu and C. C. Feng (2014). "Persistence of extracellular DNA in river sediment facilitates antibiotic resistance gene propagation." Environmental science & technology 48(1): 71-78. Marti, E., E. Variatza, J. Luis Balcazar and J. L. Balcazar (2014). "The role of aquatic ecosystems as reservoirs of antibiotic resistance." Trends in Microbiology 22(1): 36- 41. Martínez, J. L. J. (2008). "Antibiotics and antibiotic resistance genes in natural environments." Science 321(5887): 365-367. Mubedi, J. I., N. Devarajan, S. L. Faucheur, J. K. Mputu, E. K. Atibu, P. Sivalingam, K. Prabakar, P. T. Mpiana, W. Wildi and J. Poté (2013). "Effects of untreated hospital effluents on the accumulation of toxic metals in sediments of receiving system under tropical conditions: Case of South India and Democratic Republic of Congo." Chemosphere 93(6): 1070-1076. Mwanamoki, P. M., N. Devarajan, B. Niane, P. Ngelinkoto, F. Thevenon, J. W. Nlandu, P. T. Mpiana, K. Prabakar, J. I. Mubedi, C. G. Kabele, W. Wildi and J. Pote (2015). "Trace metal distributions in the sediments from river-reservoir systems: case of the Congo River and Lake Ma Vallée, Kinshasa (Democratic Republic of Congo)." Environ Sci Pollut Res Int 22(1): 586-597. Mwanamoki, P. M., N. Devarajan, F. Thevenon, E. K. Atibu, J. B. Tshibanda, P. Ngelinkoto, P. T. Mpiana, K. Prabakar, J. I. Mubedi, C. G. Kabele, W. Wildi and J. Pote (2014). 121 "Assessment of pathogenic bacteria in water and sediment from a water reservoir under tropical conditions (Lake Ma Vallee), Kinshasa Democratic Republic of Congo." Environ Monit Assess 186(10): 6821-6830. Ngelinkoto, P., F. Thevenon, N. Devarajan, N. Birane, J. Maliani, A. Buluku, D. Musibono, J. I. Mubedi and J. Poté (2014). "Trace metal pollution in aquatic sediments and some fish species from the Kwilu-Ngongo River, Democratic Republic of Congo (Bas- Congo)." Toxicol Environ Chem 96(1): 48-57. Niane, B., S. Guedron, R. Moritz, C. Cosio, P. M. Ngom, S. Guédron, N. Deverajan, H. R. Pfeifer and J. Poté (2015) "Human exposure to mercury in artisanal small-scale gold mining areas of Kedougou region, Senegal, as a function of occupational activity and fish consumption." Environ Sci Pollut R 22, 7101-7111 DOI: 10.1007/s11356-014- 3913-5. Nzolo, D. D., P. R. Mulungo and a. l. s. d. p. SIAPS (2013). Évaluation de l’utilisation des médicaments dans six hôpitaux pilotes où les CPT sont fonctionnels en République démocratique du Congo. Orsi, R. H., N. C. Stoppe, M. I. Z. Sato, T. A. T. Gomes, P. I. Prado, G. P. Manfio and L. M. M. Ottoboni (2007). "Genetic variability and pathogenicity potential of Escherichia coli isolated from recreational water reservoirs." Research in Microbiology 158(5): 420-427. Ovreas, L., L. Forney, V. Torsvik, L. Ovreås and F. L. Daae (1997). "Distribution of bacterioplankton in meromictic Lake Saelenvannet, as determined by denaturing gradient gel electrophoresis of PCR-amplified gene fragments coding for 16S rRNA." Applied and environmental microbiology 63(9): 3367-3373. Pardos, M., C. Benninghoff, L. F. De Alencastro and W. Wildi (2004) "The impact of a sewage treatment plant's effluent on sediment quality in a small bay in Lake Geneva (Switzerland–France). Part 1: Spatial distribution of contaminants and the potential for biological impacts." Lakes & Reservoirs: Research & Management 9, 41-52 DOI: 10.1111/j.1440-1770.2004.00233.x. Poirel, L. L., P. P. Nordmann and R. A. R. Bonnin (2012). "Genetic support and diversity of acquired extended spectrum b-lactamases in Gram-negative rods." Infection, genetics and evolution 12(5): 883-893. Poté, J., L. Haller, R. Kottelat, V. Sastre, P. Arpagaus and W. Wildi (2009). "Persistence and growth of faecal culturable bacterial indicators in water column and sediments of Vidy Bay, Lake Geneva, Switzerland." Journal of environmental sciences 21(1): 62-69. 122 Poté, J., L. Haller, J.-L. Loizeau, A. Garcia Bravo, V. Sastre and W. Wildi (2008). "Effects of a sewage treatment plant outlet pipe extension on the distribution of contaminants in the sediments of the Bay of Vidy, Lake Geneva, Switzerland." Bioresource Technology 99(15): 7122-7131. R Core Team (2015). R: A language and environment for statistical computing. Vienne, Austria, R Foundation for Statistical Computing. Ram, J. L., R. P. Ritchie, J. Fang, F. S. Gonzales and J. P. Selegean (2004). "Sequence-based source tracking of Escherichia coli based on genetic diversity of beta-glucuronidase." Journal of Environmental Quality 33(3): 1024-1032. Roos-Barraclough, F., N. Givelet, A. Martinez-Cortizas, M. E. Goodsite, H. Biester and W. Shotyk (2002). "An analytical protocol for the determination of total mercury concentrations in solid peat samples." Science of The Total Environment 292(1–2): 129-139. Sakan, S. M., D. S. Dordevic, D. D. Manojlovic and P. S. Predrag (2009). "Assessment of heavy metal pollutants accumulation in the Tisza river sediments." Journal of environmental management 90(11): 3382-3390. Seiler, C. and T. U. Berendonk (2012). "Heavy metal driven co-selection of antibiotic resistance in soil and water bodies impacted by agriculture and aquaculture." Frontiers in Microbiology 3. Sidrach Cardona, R. R., E. E. Marti, J. L. J. Balcazar, E. E. Becares, M. M. Hijosa Valsero and J. L. J. Balcázar (2014). "Prevalence of antibiotic-resistant fecal bacteria in a river impacted by both an antibiotic production plant and urban treated discharges." Science of the total environment 488-489: 220-227. Spindler, A., L. M. Otton, D. B. Fuentefria and G. Corcao (2012). "Beta-lactams resistance and presence of class 1 integron in Pseudomonas spp. isolated from untreated hospital effluents in Brazil." Antonie Van Leeuwenhoek 102(1): 73-81. Suzuki, S., M. Ogo, T. Koike, H. Takada and B. Newman (2015) "Sulfonamide and tetracycline resistance genes in total- and culturable-bacterial assemblages in South African aquatic environments." Front Microbiol 6 DOI: 10.3389/fmicb.2015.00796. Tacao, M., A. Moura, I. Henriques, M. Tacão and A. Correia (2014). "Co-resistance to different classes of antibiotics among ESBL-producers from aquatic systems." Water research 48: 100-107. 123 Thevenon, F., T. Adatte, W. Wildi and J. Pote (2012). "Antibiotic resistant bacteria/genes dissemination in lacustrine sediments highly increased following cultural eutrophication of Lake Geneva (Switzerland)." Chemosphere 86(5): 468-476. Thevenon, F., N. Regier, C. Benagli, M. Tonolla, T. Adatte, W. Wildi and J. Poté (2012). "Characterization of fecal indicator bacteria in sediments cores from the largest freshwater lake of Western Europe (Lake Geneva, Switzerland)." Ecotoxicology and Environmental Safety 78: 50-56. Tshibanda, J. B., N. Devarajan, N. Birane, P. M. Mwanamoki, E. K. Atibu, P. T. Mpiana, K. Prabakar, J. Mubedi Ilunga, W. Wildi and J. Poté (2014). "Microbiological and physicochemical characterization of water and sediment of an urban river: N’Djili River, Kinshasa, Democratic Republic of the Congo." Sustainability of Water Quality and Ecology 3–4: 47-54. Verlicchi, P., A. Galletti, M. Petrovic, D. Barcelo and D. Barceló (2010). "Hospital effluents as a source of emerging pollutants: An overview of micropollutants and sustainable treatment options." Journal of hydrology 389(3-4): 416-428. WHO (2014). Antimicrobial resistance: global report on surveillance 2014: 257. WHO (2015). Drinking-water World Health Organisation Xi, C., Y. Zhang, C. F. Marrs, W. Ye, C. Simon, B. Foxman and J. Nriagu (2009). "Prevalence of Antibiotic Resistance in Drinking Water Treatment and Distribution Systems." Applied and environmental microbiology 75(17): 5714-5718. Zhao, L., D. Mi, Y. Chen, L. Wang and Y. Sun (2015) "Ecological risk assessment and sources of heavy metals in sediment from Daling River basin." Environ Sci Pollut R 22, 5975- 5984 DOI: 10.1007/s11356-014-3770-2. 124 CHAPTER 5 Strong impact of anthropogenic activities on the dissemination of toxic metals, extending- spectra β-lactamases and carbapenem resistance in sub-Saharan African urban rivers Manuscript in preparation 125 Abstract The data concerning the occurrence and dissemination of toxic metals, antibiotic resistant bacteria and their resistance genes into the aquatic ecosystems of sub-Saharan African countries are quite limited. These contaminants may have major threat to human health and aquatic organisms. In this study, river sediments receiving multiple untreated urban sewage and point source hospital effluents from the city of Kinshasa (RD Congo) were analyzed to investigate the abundance and distribution of toxic metals, and antibiotic resistance genes (ARGs). ARGs liked to β-lactam resistance (blaCTX-M and blaSHV), carbapenem resistance (blaVIM, blaIMP, blaKPC, blaOXA-48 and blaNDM) and total bacterial load were quantified by targeting specific genes using quantitative PCR in total DNA extracted from sediment. Toxic metal analysis was performed using ICP-MS. The results revealed highest abundance of 16S rRNA and ARGs copy numbers in sediment samples. Strong pollution of rivers by toxic metals was found, with values (mg kg-1) of 81.85 (Cr), 5.09 (Co), 33.84 (Ni), 203.46 (Cu), 1055.92 (Zn), 324.24 (Pb) and 2.96 (Hg). Significant correlations were observed between (i) metals (except Cd and Hg) and organic matter (R>0.6, p<0.05); and (ii) ARGs (except blaNDM) and 16S rRNA (R>0.57, p<0.05). The study clearly demonstrated that several sources of pollutants contribute to the degradation of water quality and to the spread of toxic metals and ARGs in aquatic ecosystems. 126 5.1 Introduction Anthropogenic activities lead to the introduction of an increasing number of priority and emerging contaminants in adjacent aquatic systems by the inflow of effluents from various origins such as municipal and industrial treated wastewater or urban/agricultural runoff. In developing countries, where people lack of sanitation facilities and where solid and liquid wastes remain often unmanaged, the situation is particularly alarming. Effluents produced by municipalities, industries, hospitals and agricultural practices are directly discharged in the nearest aquatic system (i.e. river, lakes and seas) without prior treatment, leading to a critical accumulation of various emerging contaminant, such as toxic metals, antibiotic resistance bacteria (ARBs) and their associated antibiotic resistance genes (ARGs), in the receiving sediments (Spindler et al. 2012, Devarajan et al. 2015, Laffite et al. 2016). Contaminants accumulated in the sediment may be remobilized by turbulence processes (i.e. recreational bathing, storm events). Freshwater is a major public resource, and in developing countries where people suffer of social and economic difficulties, a large proportion of the population use highly contamination surface water, shallow wells and boreholes for irrigation, domestic and drinking purposes (Rochelle-Newall et al. 2015, Kapembo et al. 2016). Knowing that the preliminary pathway of metal accumulation in the human body and water/food borne disease are through the consumption of contaminated food and water (Miller et al. 2004, WHO 2011, WHO 2017), its consumption by local population represent a human health hazard. Beta-lactams antibiotics, which include penicillins, cephalosporins, monobactams and carbapenems, are the most used antibiotics over the world. Among β-lactams, carbapenems are by far the most effectives against many bacteria species presenting a broad spectrum of antibacterial activity. Over the time, β-lactam-hydrolysing enzymes which inactivates penicillins evolved, and extended their spectra to cephalosporins, then to ESBLs and recently to carbapenemases (Codjoe and Donkor 2017). Carbapenemases enzymes are a diverse group of enzymes belonging to the classes A, B, and D of Ambler classification. All carbapenemases enzymes share a serine residue in the active site, the ability to inactivate carbapenems antibiotics, and can be chromosomally or plasmid encoded. In term of carbapenem hydrolysis and geographical spread, the plasmid encoded enzymes KPC, VIM, IMP, NDM and OXA-48 type are described as the more effective (Poirel 2012). Actually, carbapenems are considered as the most reliable last-resort antibiotics, because it uses lead to fewer adverse effects than other last-line antibiotics. Furthermore, the involvement of carbapenem resistance in multidrug resistant bacteria (MDR) is the last step before the emergence of pandrug resistance (PDR) 127 (Meletis 2016). For these reasons, the dissemination of carbapenemase producing bacteria is a particular global concern. Many clinical studies described the widespread dissemination of carbapenem resistance in Europe, Asia and South America, however the dissemination of carbapenem resistance in African countries remain poorly understood (Dortet et al. 2014, Codjoe and Donkor 2017, Wu et al. 2019). The present study focused on 2 sub-urban rivers of the vicinity of Kinshasa (RD Congo). We investigate the presence of β-lactamase and carbapenemase resistance genes as well as the contamination by toxic metals in the river sediments aiming to assess the impact of a hospital wastewater and of a tropical rain. We hypothesized that hospital effluent may have a major impact on the accumulation and dissemination of ARGs in river and that tropical rainfall event may remobilized contaminant from the sediment and drive it to downstream location. The assessment is based on (i) quantification of β-lactam resistance genes (blaSHV, blaCTX-M, blaNDM, blaKPC, blaOXA-48, blaVIM, blaIMP) by quantitative polymerase chain reaction (qPCR) and (ii) determination of toxic metals including Cr, Co, Ni, Cu, Zn, Cd, Pb and Hg. The analysis of sediment samples was performed before and after a tropical rainfall event in order to evaluate the impact of storm event. 5.2 Material and Methods 5.2.1 Study sites and sampling procedures Two urban rivers receiving hospital effluents were selected in the vicinity of Kinshasa (RD Congo) according to our previous publications (Mubedi et al. 2013, Kilunga et al. 2016, Laffite et al. 2016). Sampling took place in December 2017. The surface sediments (0-4cm layer) were collected at the reject point of the hospital effluent (Eff) as well as 100m upstream (US) and 100 m downstream (DS) the effluent. Samples were taken before (Dry) and after (Wet) a tropical raining condition. Samples were collected in sterile polypropylene tubes and transported in an icebox at 4°C and shipped to the University of Geneva (Switzerland). 5.2.2 Sediment physico-chemical parameters: sediment grain size, total organic matter and carbonates The particle grain size was measured using a Laser Coulter® LS-100 diffractometer (Beckman Coulter, Fullerton, Ca, USA), following 5 min ultrasonic dispersal in deionized water according to the method prescribed by Loizeau et al. (1994). The sediment total organic matter 128 (OM) and carbonates (CaCO3) content were estimated by loss on ignition at 550°C and 1000°C respectively, for 1 h in a Salvis oven (Salvis AG, Emmenbrücke, Lucerne, Switzerland). 5.2.3 Toxic metal analysis in sediment samples Before being analyzed, sediment samples were lyophilized after homogenization. Toxic metals including Cr, Cu, Zn, Cd and Pb were determined by Inductively Coupled Plasma Mass Spectrometry (ICP-MS, Agilent model 7700 series) following the digestion in analytical grade 2M HNO3 at 150°C (Loizeau et al. 1994, Laffite et al. 2016). Multi-element standard solutions at different concentrations (0, 0.02, 1, 5, 20, 100, and 200 µg/L) were used for calibration. Total variation coefficients of triplicate sample measurements were under 5% and chemical blanks for the procedure were less than 2% of the sample signal. The metal concentrations of sediments were expressed in ppm (mg kg−1 dry weight of sediment). Total Hg analysis was carried out using the Atomic Absorption Spectrophotometer (AAS) for mercury determination (Advanced Mercury Analyser; AMA 254, Altec s.r.l., Czech Rep.) following the method described by E. M. Hall and Pelchat (1997) and Roos-Barraclough et al. (2002). The method is based on sample combustion, gold amalgamation, and AAS. The detection limit (3 SD blank) was 0.005mg kg−1 and the reproducibility better than 2%. 5.2.4 Geoaccumulation index (Igeo), enrichment factor (EF), and single pollution index (PI) In order to evaluate the possible environmental risks, three indexes were determined. The geoaccumulation index was used to assess the contamination of the sediment. It was calculated using the following equation: 퐶푛 퐼푔푒표 = 푙표푔 2 ( ) 1.5 퐵푛 Where Cn is the concentration of a given element in the sediment sample and Bn is the concentration of the elements in geochemical background. The constant value 1.5 in the equation considers natural fluctuations of a given substance in the environment. Seven quality classes of Igeo classified the increasing sediment contamination (Müller 1981). 129 The enrichment factor was used to distinguish anthropogenic and natural sources of trace metal using the formula: 푋 (푆푐) 퐸퐹 = 푠푎푚푝푙푒 푋 ( ) 푆푐 퐿푀푉 Where, X represents the element of interest; EFX is the enrichment factor of X, Xsample is the concentration of sample; XLMV is the concentration of the element in the reference background (Varol 2011). As RD Congo is naturally rich in various metals including toxic metals, we choose to use the Lac MaVallée (LMV) metal content as reference background (Mwanamoki et al. 2014). Indeed, La MaVallée is a small artificial lake reservoir located at ~15km of Kinshasa, surrounded by a secondary mature forest in the tropics, where the population of Kinshasa use to go for leisure and recreation. This site is protected and previous studies showed that the site was not impacted by anthropogenic activities (Mwanamoki et al. 2014, Mwanamoki et al. 2015). Finally, single pollution index (PI) was calculated for the assessment of sediment contamination level and to determine which metal represents the highest treat using the formula: 퐶푖 푃퐼 = 푆푖 Where PI is the pollution index, Ci is the measured concentration of each metal in the sediment and Si is the background value in the study. Sediments were classified in 5 classes: non polluted (PI < 1),low level of pollution (1 > PI < 2), moderate level of pollution ( 2 ≥ PI < 3), strong level of pollution (3 > PI < 5) and very strong level of pollution (PI > 5) (Tripathee et al. 2016). 5.2.5 ARGs quantitation by qPCR Total DNA from sediment samples was extracted using the PureLink™ Microbiome DNA Purification Kit (Life Technologies, Zug, Switzerland) according to manufacture’s recommendations. DNA extraction was performed with three replicates from the same sample to compensate the heterogeneity. The concentration of extracted DNA was measured using the Quant-iT™ Picogreen™ dsDNA assay Kit (Life Technologies, Zug, Switzerland). The isolated DNA was stored at – 20 °C until used. 130 The quantitation of ARGs (blaSHV, blaCTX-M, blaNDM, blaKPC, blaOXA-48, blaVIM, blaIMP), and 16S rRNA genes by qPCR was performed as previously described by Laffite et al. (2016), Bisiklis et al. (2007), Poirel et al. (2011), Swayne et al. (2011) (Table 5-1). Briefly, genes were quantitated with Eco qPCR system (Illumina, Switzerland) using SensiFAST™ SYBR® Kit (Bioline, London, UK). The following cycling conditions were applied: 2 min at 95 °C for polymerase activation; followed by 40 cycles of 95 °C for 5 s, optimal Tm for 10 s and 72 °C for 10 s. The temperature melting profile was obtained using the following conditions; 95 °C for 30 s, optimal Tm for 30 s, followed by 95 °C for 30 s. To emphasize the relative abundance of the resistance genes the concentrations of the gene copy numbers were presented as percentage of “copy number of a gene/copy number of 16S rRNA” for each sample. Table 5-1 Primer table for qPCR quantitation Gene Sequence (5' → 3') Size (pb) Tm (°C) Reference Sidrach bla GCKGCCAACTTACTTCTGACAACG TEM 247 55 Cardona et CTTTATCCGCCTCCATCCAGTCTA al. (2014) TCAGCGAAAAACACCTTG Xi et al. blaSHV 110 60 (2009) TCCCGCAGATAAATCACCA ATTCCRGGCGAYCCGCGTGATACC Fujita et al. blaCTX-M 227 62 (2011) ACCGCGATATCGTTGGTGGTGCCAT TTGGCGATCTGGTTTTCC Zheng et al. blaNDM 195 58 (2013) GGTTGATCTCCTGCTTGA GCAGCGGCAGCAGTTTGTTGATT Swayne et blaKPC 184 62 al. (2011) GTAGACGGCCAACACAATAGGTGC GCG TGG TTA AGG ATG AAC AC Poirel et al. blaOXA-48 438 55 (2011) CAT CAA GTT CAA CCC AAC CG bla GTACGCATCACCGTCGACAC VIM Bisiklis et 172 60 TGACGGGACGTATACAACCAGA al. (2007) AGACGGGACGTACACAACTAAG blaIMP AAGTTAGTCAMTTGGTTTGTGGAGC Bisiklis et 269 56 al. (2007) CAAACCACTACGTTATCTKGAGTGTG Bacterial ACTCCTACGGGAGGCAGCAG Ovreas et 197 55 16S rRNA ATTACCGCGGCTGCTGG al. (1997) 5.2.6 Statistical analysis Resistance gene copy number was normalized to bacterial 16S rRNA gene copy number of the same sample in order to analyze the relative abundance change of each resistance gene 131 in the bacterial population. All statistical analysis were done with Rstudio version 3.4.4. with a statistical significance defined as α < 0.05. In order to determine the impact of point source and rainfall on the dissemination of ARGs in the sediment, ANOVA and pairwise comparison were used. Spearman correlations between physicochemical parameters, metals and genes copy numbers on a log10 scale were used to evaluate possible relationships. A principle component analysis (PCA) using a correlation matrix was performed to allow direct comparisons. 5.3 Results 5.3.1 Sediment Physicochemical Parameters Sediment characteristics including particle grain-size, total organic matter (OM) and carbonates (CaCO3) content are presented Table 5-2. Surface sediment of river were classified as silty sandy to sand with less than 3% of clay and range between 76.98-352.4 µm (R1) and 61.59-279.1 µm (R2). The river sediment granulometry showed no significant variation according to conditions (season, river, site; p>0.05). Total organic matter and carbonates varied substantially from 1.02 to 22.81% (R1) and 1.24 to 33.16% (R2) with a significantly different distribution between rivers (p<0.05). Whereas, OM and CaCO3 distribution was not affected by the season nor the input of the studied hospital effluent (p>0.05). Table 5-2 Physicochemical parameters in sediment samples Median grain Season River Site OM (%) CaCO3 Clay (%) Silt (%) Sand (%) size (µm) Dry R1 US 22.40 32.51 1.19 33.64 65.17 130 Eff 17.40 26.28 1.30 43.39 55.31 76.98 DS 1.51 2.58 0.63 8.11 91.26 352.4 R2 US 6.89 10.29 0.56 8.61 90.83 236.5 Eff 1.23 1.91 0.00 1.06 98.94 279.1 DS 1.02 1.69 0.56 8.65 90.79 175.8 Wet R1 US 22.81 33.16 0.67 25.95 73.38 158.7 Eff 0.67 1.24 0.56 8.34 91.10 340.8 DS 17.92 26.50 1.10 29.96 68.94 165.1 R2 US 3.38 5.08 2.57 28.13 69.30 106.1 Eff 9.24 13.68 2.51 48.08 49.41 61.59 DS 1.19 1.99 0.79 10.69 88.52 160.6 US: Upstream; Eff: Effluent; DS: Downstream; R1: river; R2: River 2 132 5.3.2 Sediment metal content and toxic metal pollution assessment Trace metal concentrations in sediments are presented Table 5-3. Before the raining event, metal concentration (in mg kg-1) in sediment varied from 4.38-35.69 (Cr), 0.52-5.09 (Co), 2.31-17.92 (Ni), 13.78-145.42 (Cu), 87.20-1013.03 (Zn), 0.14-3.56 (Cd), 21.82-184.08 (Pb) and 0.08-2.52 (Hg). The concentration of Cu, Zn and Hg were above the sediment quality guidelines for the aquatic fauna and especially for the Zn, which exceed the probable effect level (CCME 1999). After the raining event, metal concentration (in mg kg-1) in sediment varied from 11.94-81.85 (Cr), 1.35-3.87 (Co), 10.80-33.84 (Ni), 34.37-203.46 (Cu), 123.87-1055.92 (Zn), 0.12-2.26 (Cd), 79.19-324.24 (Pb) and 0.07-2.96 (Hg). An increase of metal concentration in sediments was observed after the rain event for Cr, Ni, Cu and Pb, whereas for the other metals concentration can either increase or decrease depending the sampling points. Furthermore, the concentration of Cr, Cu, Zn and Hg were higher than the recommended values for sediment quality guidelines (CCME 1999). Table 5-3 Metal content analyzed by ICP-MS (mg kg-1 dry weight) in sediments according to the rain event, the sampled river and the sampling location Sampling Cr Co Ni Cu Zn Cd Pb Hg Season River point Dry R1 US 35,69 5,09 17,92 145,42 1013,03 3,56 184,08 2.15 Eff 32,70 3,14 11,38 82,82 631,60 1,57 105,19 2.52 DS 9,33 1,37 7,00 51,62 206,25 0,63 58,37 0.08 R2 US 7,07 0,93 3,19 27,18 242,67 0,32 54,92 0.10 Eff 4,38 0,54 2,35 16,38 87,20 0,14 21,82 0.11 DS 4,68 0,52 2,31 13,78 169,48 0,17 36,98 0.06 Wet R1 US 67,55 3,87 33,84 203,46 862,57 1,98 324,24 2.19 Eff 11,94 1,35 10,80 34,37 123,87 0,35 79,19 0.07 DS 43,78 5,18 18,45 138,39 1055,92 2,26 160,26 2.96 R2 US 65,76 2,73 22,57 93,27 652,18 0,12 236,57 0.07 Eff 81,85 2,37 29,44 170,70 798,04 1,22 296,60 0.76 DS 64,79 2,52 23,50 114,42 496,30 0,22 228,67 0.07 a SQGs 37,30 35,70 123,00 0,17 PELb 90,00 197,00 315,00 0,49 SQGsa: Sediment quality guidelines (mg kg-1) PELb: Probable effect level (mg kg-1) In bold: values above SQGs In bold & italic: values above PEL US: Upstream; Eff: Effluent; DS: Downstream; R1: river; R2: River 2 133 The use of the three indexes, namely geo-accumulation index (Igeo), enrichment factor (EF) and pollution index (PI) enabled to understand and evaluate the potential environmental risk by comparing metal background value to metal concentrations determined in the present study. According to the Igeo (Table 5-4), which is a measure of the sediment pollution level by environmental or organic waste as well as bio-elements, pollution by toxic metals varied from class 0 (practically unpolluted) to class 6 (extremely polluted) in the order of Cr = Ni < Co < Cu = Cd = Hg < Pb < Zn before the rain event and Ni < Cr = Co < Cd < Cu < Zn = Pb after the rain event. The Igeo index pattern varied substantially from the River 1 to the River 2 before the rain event, with a high index level in the River 1 for Zn, Cd, Pb and Hg. The difference in Igeo index between the two rivers disappeared after the rain event, with the appearance of a global increase in metal pollution. Table 5-4 Igeo values in surface sediments Igeo Dry Wet R1 R2 R1 R2 US Eff DS US Eff DS US Eff DS US Eff DS Cr 1,0 0,9 -0,9 -1,3 -2,0 -1,9 2,0 -0,5 1,3 1,9 2,2 1,9 Co 2,3 1,6 0,4 -0,2 -0,9 -1,0 1,9 0,4 2,3 1,4 1,2 1,3 Ni 0,6 0,0 -0,7 -1,9 -2,3 -2,3 1,5 -0,1 0,7 0,9 1,3 1,0 Cu 4,0 3,2 2,6 1,6 0,9 0,6 4,5 2,0 4,0 3,4 4,3 3,7 Zn 6,6 5,9 4,3 4,5 3,0 4,0 6,3 3,5 6,6 5,9 6,2 5,5 Cd 5,2 4,0 2,7 1,7 0,6 0,8 4,3 1,8 4,5 0,4 3,6 1,2 Pb 5,2 4,4 3,6 3,5 2,2 2,9 6,0 4,0 5,0 5,6 5,9 5,5 Hg 5,6 5,8 0,8 1,2 1,3 0,3 5,6 0,6 6,0 0,7 4,1 0,7 US: Upstream; Eff: Effluent; DS: Downstream; R1: River1; R2: River 2 Igeo ≤ 0 Class 0 - pratically unpolluted 0 < Igeo ≤ 1 Class 1 - unpolluted to moderately polluted 1 < Igeo ≤ 2 Class 2 - moderately polluted 2 < Igeo ≤ 3 Class 3 - moderately to heavily polluted 3 < Igeo ≤ 4 Class 4 - heavily polluted 4 < Igeo ≤ 5 Class 5 - heavily to extremely polluted 5 > Igeo Class 6 - extremely polluted 134 The enrichment factor enabled to distinguish between anthropogenic and natural sources of the trace metals. The EF values (Table 5-5) for both Cu, Zn, Cd, Pb and Hg ranged between 14.1 to 916.1 indicating a severe to extremely severe enrichment due to anthropogenic activities. Whereas Cr, Co and Ni showed moderate to severe enrichment due to anthropogenic activities. Similarly to Igeo index, EF values for all the metals increased after the rain event. Table 5-5 EF values in surface sediments EF Dry Wet R1 R2 R1 R2 US Eff DS US Eff DS US Eff DS US Eff DS Cr 8,7 11,5 10,2 7,3 6,7 7,2 20,5 10,3 9,0 32,4 11,1 73,5 Co 20,9 18,6 25,1 16,1 13,9 13,6 19,7 19,6 17,8 22,5 5,4 47,9 Ni 6,5 5,9 11,4 4,9 5,3 5,3 15,2 13,8 5,6 16,4 5,9 39,4 Cu 70,1 57,5 111,4 55,6 49,4 41,9 121,9 58,3 55,9 90,4 45,6 255,5 Zn 401,8 360,8 366,0 407,8 216,5 424,1 424,9 172,9 350,9 520,0 175,4 911,3 Cd 155,0 98,6 123,3 58,3 39,1 46,3 107,1 53,6 82,7 11,0 29,5 44,2 Pb 159,3 131,1 226,0 201,4 118,2 201,9 348,4 241,1 116,2 411,5 142,2 916,1 Hg 203,1 343,0 33,4 41,6 64,5 33,3 256,4 22,8 234,3 14,1 39,6 32,6 US: Upstream; Eff: Effluent; DS: Downstream; R1: River 1; R2: River 2 EF < 1 indicates no enrichment EF < 3 minor enrichment EF 3 - 5 moderate enrichment EF 5 - 10 moderately severe enrichment EF 10 - 25 severe enrichment EF 25 - 50 very severe enrichment EF > 50 extremely severe enrichment The pollution index (PI) allowed assessing the degree of metal contamination. The PI (Table 5-6) varied greatly according to the metal concerned but also between the two rivers and according to the impact of the rain event. Before the rain event, the PI values for Cr, Co and Ni ranged between 0.3 to 7.4 indicating mainly a low level of pollution. The PI value of Cu, Zn, Cd, Pb and Hg in the River 1 (except Hg downstream) indicates a very strong level of pollution whereas in the River 2, there was a fluctuation between low level of pollution to strong level of 135 pollution and until very strong level of pollution for Zn and Pb. After the rain episode, PI globally increases for all the elements and reaches most of time a very strong level of pollution. Table 5-6 PI values in surface sediments PI Dry Wet R1 R2 R1 R2 US Eff DS US Eff DS US Eff DS US Eff DS Cr 3,1 2,8 0,8 0,6 0,4 0,4 5,8 1,0 3,8 5,7 7,1 5,6 Co 7,4 4,6 2,0 1,3 0,8 0,8 5,6 2,0 7,5 4,0 3,4 3,7 Ni 2,3 1,5 0,9 0,4 0,3 0,3 4,3 1,4 2,4 2,9 3,8 3,0 Cu 24,8 14,1 8,8 4,6 2,8 2,3 34,6 5,9 23,6 15,9 29,1 19,5 Zn 141,8 88,4 28,9 34,0 12,2 23,7 120,7 17,3 147,8 91,3 111,7 69,5 Cd 54,7 24,2 9,7 4,9 2,2 2,6 30,4 5,4 34,8 1,9 18,8 3,4 Pb 56,2 32,1 17,8 16,8 6,7 11,3 99,0 24,2 48,9 72,2 90,6 69,8 Hg 71,7 84,0 2,6 3,5 3,6 1,9 72,9 2,3 98,7 2,5 25,2 2,5 US: Upstream; Eff: Effluent; DS: Downstream; R1: River 1; R2: River 2 PI < 1 non-pollution 1 ≤ PI < 2 low level of pollution 2 ≤ PI < 3 moderate level of pollution 3 ≤ PI < 5 strong level of pollution PI 5 very strong level of pollution 5.3.3 Abundance of antibiotic resistance genes in urban rivers Β-lactam and carbapenem resistance genes (except blaNDM) were detectable in quantifiable amount in all the sampling sites (Figure 5-1). Prior the rain event, ARGs abundance ranges were 7.11x10-4-1.36x10-2, 2.08x10-3-6.81x10-1, 1.43x10-4-4.68x10-3, 9.08x10-7-4.41x10- 2, 1.40x10-3-2.11x10-2, 5.21x10-4-6.85x10-2, 2.36x10-3-1.17x10-2 ARG copies /16S rRNA gene for blaCTX-M, blaIMP, blaKPC, blaNDM, blaOXA-48, blaSHV and blaVIM respectively. Apart few exceptions, no significant ARGs enrichment was found along the river (p>0.05). An enrichment increased in blaCTX-M and blaIMP (2 times and 4 times, respectively) was only observed at the reject point of the hospital effluent in the River 2 before the rain event (p < 0.01). After the rain event, ARGs abundance ranges were 1.02x10-3-1.68x10-2, 1.36x10-3-9.56x10-2, 1.54x10-4- 4.26x10-3, 136 ARG copies /16S rRNA gene for blaCTX-M, blaIMP, blaKPC, blaNDM, blaOXA-48, blaSHV and blaVIM respectively. Only the gene blaOXA-48, exhibited a significant enrichment (p<0.05) along the river. The other studied resistance genes do not show any enrichment. Interestingly, when considering ARGs abundances (in ARG copies per g of dry sediment (DS)) (Figure 5-2), significant differences in ARGs abundance can be seen along the river. Figure 5-1 Normalized ARGs copy number Figure 5-2 ARGs copy number / g DS detected in hospital receiving systems at detected in hospital receiving systems at each sampling point each sampling point. US, upstream; RP, reject point; DS, downstream. 5.3.4 Correlation between chemical and biological parameters Metal concentrations (except Cd and Hg), OM and silt were, for the most-part, significantly positively correlated (0.62 < R < 1, p<0.05) (Table 5-7). ARGs (except blaNDM), 16S rRNA gene and median grain size were significantly positively correlated (0.57 < R< 1, p<0.05). Figure 5-3 represent the results of PCA analysis for the studied contaminants and physico-chemical parameters. The PCA’s first component corresponded to 50.6% of the total 137 variance and was mainly explained by the OM and metal contamination and by the granulometric characteristics. The PCA’s second component described 25.6% of the total variance and was mainly explained by the absence of ARGs contamination. Inspection of the Figure 5-4 shows that samples cannot be separately clustered according to the season, the river nor the sampling site. However, opposite trends can be seen. Firstly, river sediments are more affected by metal contamination after rain event than during the dry period. And secondly, the River 2 is more affected by ARGs contamination then the River 1. On the contrary, the rain event does not affect ARGs deposition in the river and rivers are similarly affected by metal contamination. To finish, sampling location did not vary substantially in ARGs deposition nor metal contamination. Figure 5-3 Correlation circle for the studied Figure 5-4 Graph of individuals of the PCA parameters. Principle component is the x results with correlation circles axis, with the second component being the y axis. 138 Table 5-7 Spearman correlation Median blaCTX M blaIMP blaKPC blaNDM blaOXA.48 blaSHV blaVIM CaCO3 Cd Clay Co Cr Cu Hg Grain Ni OM Pb Sand Silt Zn Size blaCTX.M 1,00 0,74 0,44 -0,23 -0,32 0,17 0,23 -0,33 -0,50 0,15 -0,24 0,01 -0,16 -0,16 -0,31 -0,05 -0,30 -0,07 -0,17 0,17 -0,13 blaIMP 1,00 0,34 -0,30 -0,41 0,13 0,04 -0,25 -0,50 -0,11 -0,28 0,07 -0,08 -0,08 -0,12 0,09 -0,24 -0,01 0,06 -0,06 -0,14 blaKPC 1,00 -0,56 0,32 0,26 0,71 -0,24 0,00 0,17 -0,04 0,18 0,04 -0,16 -0,27 0,12 -0,30 0,09 -0,38 0,38 0,08 blaNDM 1,00 -0,20 0,03 -0,39 -0,43 -0,20 -0,73 -0,54 -0,73 -0,58 -0,32 0,69 -0,69 -0,36 -0,65 0,73 -0,73 -0,67 blaOXA-48 1,00 0,10 0,79 0,25 0,73 0,24 0,44 0,33 0,44 0,34 -0,02 0,30 0,22 0,27 -0,35 0,35 0,39 blaSHV 1,00 0,26 -0,26 -0,06 -0,34 -0,31 -0,38 -0,31 0,07 0,36 -0,34 -0,23 -0,45 0,27 -0,27 -0,29 blaVIM 1,00 -0,02 0,41 0,29 0,31 0,29 0,30 0,21 -0,17 0,27 -0,06 0,20 -0,41 0,41 0,24 CaCO3 1,00 0,77 0,59 0,82 0,57 0,78 0,85 -0,59 0,59 0,99 0,60 -0,69 0,69 0,89 Cd 1,00 0,34 0,72 0,36 0,66 0,76 -0,28 0,39 0,76 0,38 -0,55 0,55 0,70 Clay 1,00 0,71 0,79 0,71 0,39 -0,85 0,71 0,55 0,76 -0,86 0,86 0,76 Co 1,00 0,69 0,83 0,71 -0,56 0,71 0,78 0,71 -0,69 0,69 0,90 Cr 1,00 0,91 0,33 -0,73 0,98 0,51 0,99 -0,71 0,71 0,76 Cu 1,00 0,59 -0,65 0,93 0,73 0,92 -0,71 0,71 0,87 Hg 1,00 -0,42 0,39 0,87 0,32 -0,55 0,55 0,68 Median Grain Size 1,00 -0,66 -0,57 -0,72 0,92 -0,92 -0,68 Ni 1,00 0,53 0,98 -0,64 0,64 0,73 OM 1,00 0,55 -0,65 0,65 0,85 Pb 1,00 -0,69 0,69 0,76 Sand 1,00 -1,00 -0,81 Silt 1,00 0,81 Zn 1,00 139 5.4 Discussion The aim of this study was to characterize the chemical and ARGs profile of two sub-saharan urban rivers under strong anthropogenic pressure. These rivers are used for many activities, including domestic purposes, water supply and population bathing and received different wastes linked to anthropogenic activities, such as industrial and urban untreated wastes, open defection, runoff of wastewater from septic tanks and hospital wastewater discharge (Kilunga et al. 2016). We assume that hospital waste constitute a specific category of waste (Verlicchi et al. 2010). However, hospital wastewater may constitute an important contributor in terms of ARGs abundance, particularly in the case of carbapenems resistance genes, which are less available in African countries (Tadesse et al. 2017). In this context, we choose two rivers receiving an hospital wastewater outlet pipe as a point source. We identified multiple contaminants (i.e. OM, metals, ARGs) in all sites during two seasons. Analysis of metals in river sediments highlighted the high degree of contamination whatever the studied river and the sampling point. In the city of Kinshasa, many river banks are used as uncontrolled landfills, which could explain the presence of high concentrations of metals in the sediments (Mavakala et al. 2016). Furthermore, artisanal activities such as tannery may also contribute to the dissemination of metals in adjacent aquatic system. The evaluation of the potential deleterious effects towards benthic fauna was performed based on sediment quality guidelines (SQGs) and the probable effect level (PEL), a contaminant level that is likely to have an effect on biota (CCME 1999, Laffite et al. 2016). According to SQGs and PEL, the concentration of Cr, Cu, Zn and Hg in sediment may present a potential toxic effect on indigenous flora and fauna. Furthermore, most of the Zn values exceed up to 4 times PEL recommendations, thus highlighting the toxicity of urban rivers. Igeo factor enabled us to determine if the soil contamination by environmental or organic waste is due to natural fluctuation of the given substance (Müller 1981). Calculated Igeo values highlight that soils were extremely polluted by Zn and Pb and in a less extent by Cu, Cd and Hg. However, it is well known that the soil in DR Congo is naturally enriched by numerous metals (Atibu et al. 2016) and the natural background needs to be taken in account. According to our previous findings, we used metal concentration measured in the sediment of the Lac MaVallée in order to calculate the Enrichment Factor index. Indeed, RD Congo soil is naturally rich in many metals and Lac MaVallée was assessed as not subjected to anthropogenic metal contamination. EF index indicated that urban rivers are severely enriched due to anthropogenic activities and 140 cannot be attributed to geochemical background. This study highlights the high level of pollution (PI > 5) in the river, mainly due to Zn and Pb, and in a less extend by Cu, Cd and Hg. Antibiotic resistance was detected in quantifiable amount in all the samples and during all seasons (except blaNDM, after the rain event). The detection of all tested ARGs including upstream the wastewater point source highlighted the background pollution of urban rivers in the city of Kinshasa. Similar observations have been previously performed in similar environment under tropical conditions (Devarajan et al. 2016, Laffite et al. 2016). However, relative ARGs abundances were generally lower than those found in the present study, thus suggesting a potential enrichment in ARGs over the time. Furthermore, similar relative ARGs abundance data have been found in studies focussing on WWTP effluents and rivers receiving WWTP output in industrialized countries (Marti et al. 2013, Devarajan et al. 2015), meaning that freshwater system in the two study sites are as enriched in ARGs as wastewater. Interestingly, ARGs abundance are strongly and positively correlated to 16S rRNA abundance, leading to an absence of ARGs enrichment over the time and sampling site. This result corroborates the hypothesis that ARGs abundance in anthropogenic impacted environment is mainly explained by fecal matter deposition and not by selection or HGT (Karkman et al. 2019). The results clearly demonstrate the accumulation of ARGs in river receiving system under tropical condition. Additionally, in many sub-saharan countries, no policies nor management tools to facilitate the urban water treatment are implemented. The lack of sanitation facilities may explain the global dissemination of biological contaminants in urban receiving system, through the use of contamination for domestic purposes including water supply, bathing and irrigation of vegetal products. The presence of ESBL in urban river systems is particularly alarming because ESBL are usually cross-resistant to other classes of antimicrobial agents (Tacao et al. 2014). However, detection of ESBL genes has become common on surface water (Zurfluh et al. 2013, Kittinger et al. 2016, Laffite et al. 2016, Kürekci et al. 2017). It is even more alarming to detect carbapenemases. The occurrence of the carbapenemases KPC, OXA-48 and VIM has been already described in European wastewater, river and lakes (Zurfluh et al. 2013, Galler et al. 2014) as well as other part of the world such as United states (Aubron et al. 2005) and China (Chen et al. 2010). In Africa, the carbapenem situation is less monitored due to the absence of active surveillance. However, some countries such as Morocco, Kenya and South Africa have reported NDM-1 as the most dominant carbapenemase gene (Poirel et al. 2011, Poirel et al. 2011), indicating that the global dissemination of carbapenemases in the environment is currently ongoing. 141 Considering the predominance of irrational use of antibiotics and the use of counterfeit drugs in many developing countries, the presence of carbapenemase in urban river and therefore the risk of CRE acquisition appear critical for human health. Chemical and biological pollutants showed different spatiotemporal trends. On one hand, ARGs variation was only explained by the sampling location (River 1 and River 2), thus enhancing the hypothesis that the faecal contamination is the main responsible of ARGs dynamics in anthropogenic impacted rivers (Karkman et al. 2019). Indeed, anthropogenic impact on urban river may vary according to the local population (i.e. access to sanitation facilities, septic tanks along the riverbank, open defection practices…), thus leading to different spatial distribution of ARGs. Previous study reported significant differences in gene abundance according to the seasonality related to the leaching of sediment during the wet season (Knapp et al. 2012). In our study, antibiotic resistance gene abundance may be due to multiple non-point source and by the continuous discharge of biological contamination of the river. The continuous discharge of biological contaminant along the river stream can be therefore the major reason of the absence of seasonal effect on ARGs distribution. On the other hand, metal contamination exhibits a temporal trend according to the rain event impact. Indeed, during the rain event, metal contaminant bounded upstream the sampling location in the river on organic matter and fine inorganic particle (i.e. clay and silts) may be resuspended and then leached along the river (Wildi et al. 2004). Interestingly, no significant trend in ARGs nor metal enrichment were observed among sampling sites (i.e. US, Eff and DS), suggesting that metal and ARGs are widely and equally distributed along the river due to anthropogenic influence. 5.5 Conclusion The aim of this study was to investigate the abundance and dissemination of ARGs and metals in urban river receiving non-point sources under tropical conditions. Results indicate that urban rivers were heavily polluted by both ARGs and metals as a results of anthropogenic activities and may represent potential risk for both aquatic fauna and flora, and human health through the use of river water for domestic and agricultural purposes. Furthermore, the study highlighted the alarming spread of ESBL and carbapenemases in the urban settings. In this context, authors recommend the prudence and regulation for the use of antimicrobials for both human and animal medicine, to limit the spread of ARGs and ARBs in the environment and the urgent need of access to safe drinking water. 142 References Atibu, E. K., N. Devarajan, A. Laffite, G. Giuliani, J. A. Salumu, R. C. Muteb, C. K. Mulaji, J.-P. Otamonga, V. Elongo, P. T. Mpiana and J. Poté (2016). "Assessment of trace metal and rare earth elements contamination in rivers around abandoned and active mine areas. The case of Lubumbashi River and Tshamilemba Canal, Katanga, Democratic Republic of the Congo." Chemie der Erde - Geochemistry 76(3): 353-362. Aubron, C., L. Poirel, R. J. Ash and P. Nordmann (2005). "Carbapenemase-producing Enterobacteriaceae, U.S. rivers." Emerging infectious diseases 11(2): 260-264. Bisiklis, A., F. Papageorgiou, F. Frantzidou and S. Alexiou Daniel (2007). "Specific detection of blaVIM and blaIMP metallo-beta-lactamase genes in a single real-time PCR." Clinical microbiology and infection 13(12): 1201-1203. CCME (1999). Recommendation canadiennes pour la qualité des sédiments. Chen, H., W. Shu, X. Chang, J.-a. Chen, Y. Guo and Y. Tan (2010). "The profile of antibiotics resistance and integrons of extended-spectrum beta-lactamase producing thermotolerant coliforms isolated from the Yangtze River basin in Chongqing." Environmental Pollution 158(7): 2459-2464. Codjoe, F. S. and E. S. Donkor (2017). "Carbapenem Resistance: A Review." Med Sci (Basel) 6(1). Devarajan, N., A. Laffite, N. D. Graham, M. Meijer, K. Prabakar, J. I. Mubedi, V. Elongo, P. T. Mpiana, B. W. Ibelings, W. Wildi and J. Pote (2015). "Accumulation of clinically relevant antibiotic-resistance genes, bacterial load, and metals in freshwater lake sediments in Central Europe." Environ Sci Technol 49(11): 6528-6537. Devarajan, N., A. Laffite, C. K. Mulaji, J.-P. Otamonga, P. T. Mpiana, J. I. Mubedi, K. Prabakar, B. W. Ibelings and J. Poté (2016). "Occurrence of Antibiotic Resistance Genes and Bacterial Markers in a Tropical River Receiving Hospital and Urban Wastewaters." PLoS One 11(2): e0149211-e0149211. Devarajan, N., A. Laffite, P. Ngelikoto, V. Elongo, K. Prabakar, J. Mubedi, P. M. Piana, W. Wildi and J. Poté (2015). "Hospital and urban effluent waters as a source of accumulation of toxic metals in the sediment receiving system of the Cauvery River, Tiruchirappalli, Tamil Nadu, India." Environmental Science and Pollution Research 22(17): 12941-12950. Dortet, L. L., L. L. Poirel and P. P. Nordmann (2014). "Worldwide Dissemination of the NDM- Type Carbapenemases in Gram-Negative Bacteria." BioMed Research International. 143 E. M. Hall, G. and P. Pelchat (1997). "Evaluation of a Direct Solid Sampling Atomic Absorption Spectrometer for the Trace Determination of Mercury in Geological Samples." Analyst 122(9): 921-924. Fujita, S., K. Yosizaki, T. Ogushi, K. Uechi, Y. Takemori and Y. Senda (2011). "Rapid Identification of Gram-Negative Bacteria with and without CTX-M Extended- Spectrum -Lactamase from Positive Blood Culture Bottles by PCR Followed by Microchip Gel Electrophoresis." Journal of clinical microbiology 49(4): 1483-1488. Galler, H., G. Feierl, C. Petternel, F. F. Reinthaler, D. Haas, A. J. Grisold, J. Luxner and G. Zarfel (2014). "KPC-2 and OXA-48 carbapenemase-harbouring Enterobacteriaceae detected in an Austrian wastewater treatment plant." Clinical Microbiology and Infection 20(2): O132-O134. Kapembo, M. L., A. Laffite, M. K. Bokolo, A. L. Mbanga, M. M. Maya-Vangua, J.-P. Otamonga, C. K. Mulaji, P. T. Mpiana, W. Wildi and J. Poté (2016). "Evaluation of Water Quality from Suburban Shallow Wells Under Tropical Conditions According to the Seasonal Variation, Bumbu, Kinshasa, Democratic Republic of the Congo." Exposure & health 8(4): 487-496. Karkman, A., K. Parnanen and D. G. J. Larsson (2019). "Fecal pollution can explain antibiotic resistance gene abundances in anthropogenically impacted environments." Nat Commun 10(1): 80. Kilunga, P. I., J. M. Kayembe, A. Laffite, F. Thevenon, N. Devarajan, C. K. Mulaji, J. I. Mubedi, Z. G. Yav, J.-P. Otamonga, P. T. Mpiana and J. Poté (2016). "The impact of hospital and urban wastewaters on the bacteriological contamination of the water resources in Kinshasa, Democratic Republic of Congo." Journal of environmental science and health. Part A, Toxic/hazardous substances & environmental engineering 51(12): 1034-1042. Kittinger, C., M. Lipp, B. Folli, A. Kirschner, R. Baumert, H. Galler, A. J. Grisold, J. Luxner, M. Weissenbacher, A. H. Farnleitner, G. Zarfel and R. Schuch (2016). "Enterobacteriaceae Isolated from the River Danube: Antibiotic Resistances, with a Focus on the Presence of ESBL and Carbapenemases." PLoS ONE 11(11): e0165820- e0165820. Knapp, C. W., L. Lima, S. Olivares-Rieumont, E. Bowen, D. Werner and D. W. Graham (2012). "Seasonal variations in antibiotic resistance gene transport in the Almendares River, Havana, Cuba." Frontiers in Microbiology 3. 144 Kürekci, C., M. Aydin, M. Yipel, M. Katouli and A. Gündoğdu (2017). "Characterization of extended spectrum β-lactamase (ESBL)-producing Escherichia coli in Asi (Orontes) River in Turkey." Journal of Water and Health 15(5): 788-798. Laffite, A., P. I. Kilunga, J. M. Kayembe, N. Devarajan, C. K. Mulaji, G. Giuliani, V. I. Slaveykova and J. Poté (2016). "Hospital Effluents Are One of Several Sources of Metal, Antibiotic Resistance Genes, and Bacterial Markers Disseminated in Sub- Saharan Urban Rivers." Frontiers in microbiology 7: 1128-1128. Loizeau, J. L., D. Arbouille, S. Santiago and J. P. Vernet (1994). "Evaluation of a wide range laser diffraction grain size analyser for use with sediments." Sedimentology 41(2): 353-361. Marti, E., J. Jofre and J. L. Balcazar (2013). "Prevalence of Antibiotic Resistance Genes and Bacterial Community Composition in a River Influenced by a Wastewater Treatment Plant." Plos One 8(10): 8. Mavakala, B. K., S. Le Faucheur, C. K. Mulaji, A. Laffite, N. Devarajan, E. M. Biey, G. Giuliani, J.-P. Otamonga, P. Kabatusuila, P. T. Mpiana and J. Pote-Wembonyama (2016). "Leachates draining from controlled municipal solid waste landfill: Detailed geochemical characterization and toxicity tests." Waste Management 55. Meletis, G. (2016). "Carbapenem resistance: overview of the problem and future perspectives." Therapeutic Advances in Infectious Disease 3(1): 15-21. Miller, J. R., K. A. Hudson-Edwards, P. J. Lechler, D. Preston and M. G. Macklin (2004). "Heavy metal contamination of water, soil and produce within riverine communities of the Rı́o Pilcomayo basin, Bolivia." Science of The Total Environment 320(2): 189- 209. Mubedi, J. I., N. Devarajan, S. L. Faucheur, J. K. Mputu, E. K. Atibu, P. Sivalingam, K. Prabakar, P. T. Mpiana, W. Wildi and J. Poté (2013). "Effects of untreated hospital effluents on the accumulation of toxic metals in sediments of receiving system under tropical conditions: Case of South India and Democratic Republic of Congo." Chemosphere 93(6): 1070-1076. Müller, G. (1981). "Die Schwermetallbelastung des Neckars und seiner nebenflüsser: Eine Bestandsaufnahme." Chemiker Zeitung 105: 157-164. Mwanamoki, P. M., N. Devarajan, B. Niane, P. Ngelinkoto, F. Thevenon, J. W. Nlandu, P. T. Mpiana, K. Prabakar, J. I. Mubedi, C. G. Kabele, W. Wildi and J. Pote (2015). "Trace metal distributions in the sediments from river-reservoir systems: case of the Congo 145 River and Lake Ma Vallée, Kinshasa (Democratic Republic of Congo)." Environ Sci Pollut Res Int 22(1): 586-597. Mwanamoki, P. M., N. Devarajan, F. Thevenon, N. Birane, L. F. de Alencastro, D. Grandjean, P. T. Mpiana, K. Prabakar, J. I. Mubedi, C. G. Kabele, W. Wildi and J. Pote (2014). "Trace metals and persistent organic pollutants in sediments from river-reservoir systems in Democratic Republic of Congo (DRC): Spatial distribution and potential ecotoxicological effects." Chemosphere 111: 485-492. Ovreas, L., L. Forney, V. Torsvik, L. Ovreås and F. L. Daae (1997). "Distribution of bacterioplankton in meromictic Lake Saelenvannet, as determined by denaturing gradient gel electrophoresis of PCR-amplified gene fragments coding for 16S rRNA." Applied and environmental microbiology 63(9): 3367-3373. Poirel, L., A. Benouda, C. Hays and P. Nordmann (2011). "Emergence of NDM-1-producing Klebsiella pneumoniae in Morocco." J Antimicrob Chemother 66(12): 2781-2783. Poirel, L., G. Revathi, S. Bernabeu and P. Nordmann (2011). "Detection of NDM-1-producing Klebsiella pneumoniae in Kenya." Antimicrob Agents Chemother 55(2): 934-936. Poirel, L. L., T. R. T. Walsh, V. V. Cuvillier and P. P. Nordmann (2011). "Multiplex PCR for detection of acquired carbapenemase genes." Diagnostic microbiology and infectious disease 70(1): 119-123. Rochelle-Newall, E. J., H. T. M. Nguyen, Q. T. P. Le, O. Sengtaheuanghoung and O. Ribolzi (2015). "A short review of fecal indicator bacteria in tropical aquatic ecosystems: knowledge gaps and future directions." Frontiers in Microbiology 6. Roos-Barraclough, F., N. Givelet, A. Martinez-Cortizas, M. E. Goodsite, H. Biester and W. Shotyk (2002). "An analytical protocol for the determination of total mercury concentrations in solid peat samples." Science of The Total Environment 292(1–2): 129-139. Sakan, S. M., D. S. Dordevic, D. D. Manojlovic and P. S. Predrag (2009). "Assessment of heavy metal pollutants accumulation in the Tisza river sediments." Journal of environmental management 90(11): 3382-3390. Sidrach Cardona, R., E. Marti, L. J. Balcazar, E. Becares, M. Hijosa Valsero and L. J. Balcázar (2014). "Prevalence of antibiotic-resistant fecal bacteria in a river impacted by both an antibiotic production plant and urban treated discharges." Science of the total environment 488-489: 220-227. 146 Spindler, A., L. M. Otton, D. B. Fuentefria and G. Corcao (2012). "Beta-lactams resistance and presence of class 1 integron in Pseudomonas spp. isolated from untreated hospital effluents in Brazil." Antonie Van Leeuwenhoek 102(1): 73-81. Swayne, R. L., H. A. Ludlam, V. G. Shet, N. Woodford and M. D. Curran (2011). "Real-time TaqMan PCR for rapid detection of genes encoding five types of non-metallo- (class A and D) carbapenemases in Enterobacteriaceae." International Journal of Antimicrobial Agents 38(1): 35-38. Tacao, M., A. Moura, I. Henriques, M. Tacão and A. Correia (2014). "Co-resistance to different classes of antibiotics among ESBL-producers from aquatic systems." Water research 48: 100-107. Tadesse, B. T., E. A. Ashley, S. Ongarello, J. Havumaki, M. Wijegoonewardena, I. J. González and S. Dittrich (2017). "Antimicrobial resistance in Africa: a systematic review." BMC Infectious Diseases 17(1): 616-616. Tripathee, L., S. Kang, D. Rupakheti, Q. Zhang, R. M. Bajracharya, C. M. Sharma, J. Huang, A. Gyawali, R. Paudyal and M. Sillanpää (2016). "Spatial distribution, sources and risk assessment of potentially toxic trace elements and rare earth elements in soils of the Langtang Himalaya, Nepal." Environmental Earth Sciences 75(19): 1332. Varol, M. (2011). "Assessment of heavy metal contamination in sediments of the Tigris River (Turkey) using pollution indices and multivariate statistical techniques." Journal of Hazardous Materials 195(0): 355-364. Verlicchi, P., A. Galletti, M. Petrovic, D. Barcelo and D. Barceló (2010). "Hospital effluents as a source of emerging pollutants: An overview of micropollutants and sustainable treatment options." Journal of hydrology 389(3-4): 416-428. WHO (2011). Guidelines for drinking-water quality, fourth edition. WHO. (2017). "Diarrhoeal disease." 26/02/2019, from https://www.who.int/news-room/fact- sheets/detail/diarrhoeal-disease. Wildi, W., J. Dominik, J.-L. Loizeau, R. L. Thomas, P.-Y. Favarger, L. Haller, A. Perroud and C. Peytremann (2004). "River, reservoir and lake sediment contamination by heavy metals downstream from urban areas of Switzerland." Lakes & Reservoirs: Research & Management 9(1): 75-87. Wu, W., Y. Feng, G. Tang, F. Qiao, A. McNally and Z. Zong (2019). "NDM Metallo-β- Lactamases and Their Bacterial Producers in Health Care Settings." Clinical Microbiology Reviews 32(2): e00115-00118. 147 Xi, C., Y. Zhang, C. F. Marrs, W. Ye, C. Simon, B. Foxman and J. Nriagu (2009). "Prevalence of Antibiotic Resistance in Drinking Water Treatment and Distribution Systems." Applied and environmental microbiology 75(17): 5714-5718. Zheng, F., J. Sun, C. Cheng and Y. Rui (2013). "The establishment of a duplex real-time PCR assay for rapid and simultaneous detection of blaNDM and blaKPC genes in bacteria." Annals of Clinical Microbiology and Antimicrobials 12(1): 30. Zurfluh, K., H. Hachler, M. Nuesch-Inderbinen and R. Stephan (2013). "Characteristics of Extended-Spectrum β-Lactamase- and Carbapenemase-Producing Enterobacteriaceae Isolates from Rivers and Lakes in Switzerland." Applied and environmental microbiology 79(9): 3021-3026. 148 CHAPTER 6 Virulence and multidrug resistance in 3G β- lactam resistant Escherichia coli from sub- Saharan African urban rivers Manuscript in preparation 149 Abstract The emergence and spread of antimicrobial resistance is a global health concern. Nevertheless, the occurrence and dissemination of multidrug resistance (MDR) and antibiotic resistance genes (ARGs) in African freshwater ecosystems are largely unknown. In the present study, 150 cefotaxim resistant isolates were isolated from five rivers receiving hospital effluents located in Kinshasa, the capital of Democratic Republic of the Congo. The isolates were tested against 12 antibiotic classes. The presence of antibiotic resistance genes (ARGs: blaTEM, blaSHV, blaCTX-M and sul genes), virulence factors (VFs) (i.e. LT1, ST1, stx1, stx2, hlyA, aggR and ipaH) and pathogenic islands (PAIs) (i.e. She, HPI, LEE, SHI-2, Tia, EspC and O-islands) were determined by end-point PCR. High level of resistance was observed in all studied strains. However, no relationship between hospital effluent release and ARGs dissemination was found. VFs and PAIs linked to diarrheic E. coli were also detected in 15.3% and 28.2% of strains, respectively. The results revealed the endemicity of MDR in tropical urban rivers receiving untreated wastewater. The high level of MDR strains, virulence factors and pathogenic islands in the studied strains highlighted the threat to public health linked to urban river water use. 150 6.1 Introduction Millenium Developpement Goals target for improved access to safe drinking water (target 7) was globally achieved, however a large population in the world still lack sustainable access to safe drinking water and basic sanitation. In Sub-Saharan Africa, 32% of the population still depends on unsafe water sources such as river or boreholes (UNICEF and WHO 2015). The use of unimproved water sources for domestic purposes and recreational activities is responsible of nearly 1.7 billion cases of childhood diarrhoeal disease and kill around 525’000 children under five each year (WHO 2017). Furthermore, due to the lack of sanitation facilities and effective water treatment infrastructures, rivers are the final receptor of unmanaged waste and are heavily polluted by the large and unregulated discharge of a large variety of contaminant, including untreated domestic, hospital and industrial waste, as well as human sewage and animal waste (Laffite et al. 2016). The resulting faecal contamination is a known pathway for the waterborne transmission of enteric pathogens and the use of surface water for domestic purposes without adequate sanitation will increase the risk of waterborne disease (Gonzales-Siles and Sjoling 2016, Park et al. 2018). For example, the actual endemicity of cholera (Vibrio cholera), a disease transmitted through faecal contaminated water, affects 47 countries across the globe and results in around 2.9 million cases and 95’000 death per year (WHO 2017). Due to strong variability, the detection of waterborne pathogens is expensive and complex (Ferguson et al. 2012). Environmental agencies and health organization recommend the use of indicators of faecal contamination to monitor the water quality and health hazard (WHO 2017). Escherichia coli is the predominant bacteria of the human colonic flora. E. coli is usually a harmless and normal inhabitant of the gastrointestinal tract, however several highly adapted clones have evolved the ability to cause a broad spectrum of gastrointestinal disease (Intestinal Pathogenic E. coli or IPEC) including diarrhoeal disease, bacteraemia, sepsis and meningitis by the acquisition of virulence properties (Rivas 2015). Other E. coli strains cause infections outside the gastrointestinal tract (extraintestinal pathogenic E. coli or ExPEC) such as urinary tract infections (UTI). IPEC are now described into six categories according to their virulence mechanism: enteropathogenic E. coli (EPEC), enterohemorrhagic E. coli (EHEC), enterotoxigenic E. coli (ETEC), enteroaggregative E. coli (EAEC), enteroinvasive E. coli (EIEC) and diffusely adherent E. coli (DAEC) (Nataro and Kaper 1998). The virulence attributes are frequently found in specific regions called pathogenicity islands (PAIs) which encode various virulence factors such as protein secretion systems, host invasion factors, iron 151 uptake and toxins (Rivas 2015). These PAIs are frequently flanked by repeated sequences and can carry many fragments of other mobile and accessory genetics elements (plasmids, insertion sequences) (Naderi et al. 2016). Furthermore, virulence factors can also be encoded on mobile genetic elements such as the plasmid pO157 for many STEC and can be easily mobilized into different strains to create novel combinations of virulence factors (Rivas 2015). Diseases caused by pathogenic E. coli often require an antimicrobial therapy and antibiotic resistant strains usually cause longer and more severe illness than antibiotic- susceptible strains (Shahrani et al. 2014). Despite that antimicrobial resistance and bacterial virulence evolved on different timescales, they share common characteristics: their processes are necessary to survive under adverse conditions, virulence and resistance are mediated by common processes (i.e. involvement of efflux pumps, porins, cell wall alteration, activation or repression of various genes), and they determinant are transmitted between species or genera by horizontal gene transfer (HGT) (Beceiro et al. 2013). Those facts, easily explained the fact that virulence and antimicrobial resistance are often associated and that this association can increase the fitness of certain species in some environment, thus leading to treatment failures (Beceiro et al. 2013). β-lactams antibiotics, including the subgroups of penicillins, cephalosporins and carbapenems, inhibit the biosynthesis of peptidoglycan by competing with the natural substrate for penicillin-binding proteins. Due to their low toxicity, β-lactams are the most prescribed and consumed antibiotics in human medicine in most countries (Kümmerer 2009, WHO 2018). Resistance to β-lactam antibiotics is mediated by a large diversity of enzymes categorized according their hydrolytic profile and on the inactivation properties of the β-lactamases inhibitors (Bush and Jacoby 2010). Extending-spectrum β-lactamases (ESBL) are able to hydrolyse penicillins, cephalosporins and aztreonam, thus limiting therapeutic options in case of infection. β-lactamases as well as other ARGs are generally located on mobile genetic elements (MGEs) such as plasmids, integrons and transposons, and being subsequently disseminated by horizontal gene transfer (HGT) (Fondi and Fani 2010, Poirel et al. 2012). The aim of the present study is to evaluate the dissemination of multidrug resistance (MDR) and antibiotic resistance genes (ARGs) in urban rivers under tropical conditions. More specifically, we evaluate the occurrence and distribution of virulence genes, pathogenic islands, phylogenetic grouping and multidrug resistance in a collection of ESBL-producing E. coli isolated from five urban river under tropical conditions. 152 6.2 Material and Methods 6.2.1 Study site and sampling settings Kinshasa, the capital and the largest city in the Democratic Republic of the Congo (DR Congo) is the 17th largest urban area in the world with more than 12’000’000 inhabitants in 2017 and covering 9’965 km² (Figure 6-1). The weather is hot and humid with an average temperature of 21-30°C. From October to May, Kinshasa is affected by the rainy season, with mean precipitation of 167 mm per month. Figure 6-1 Localization of the sampling site in the province of Kinshasa, Republic Democratic of Congo (adapted from Google Earth) The sampling campaign took place in January 2016. Five urban rivers (R1 to R5) affected to hospital wastewater discharge were selected in the vicinity of Kinshasa (Democratic Republic of the Congo). Assuming that hospital waste constitute a specific category of waste in term of infectivity and toxicity (Verlicchi et al. 2010), thus leading to an important contribution on the development of the AMR in receiving rivers, the sampling sites were chosen on the basis of (i) hospital size and innovativity, (ii) absence of wastewater treatment previous discharge and (iii) absence of industries next to the sampling location. Water samples were collected in sterile PP bottles across a distance of 400m from the hospital effluent pipe and sent to the laboratory of the Department F.-A. Forel in an icebox within 48h hours of the collection and processed immediately for microbiological analysis. 153 6.2.2 Escherichia coli isolation and antimicrobial susceptibility testing To assess the impact of hospital wastewater discharge on the dissemination of β- lactamase strain, a collection of β-lactam resistant E. coli consisting on 150 isolates was carried out. After culture on TBX (Biolife Italiana, Milan, Italy) supplemented with 2 µg ml-1 cefotaxime (Sigma Aldriich, Switzerland), 6 random colony per sampling site with a color corresponding to E. coli (blue) were chosen for further analysis. E. coli isolates were confirmed by matrix-assisted laser desorption/ionization time-of-flight mass spectrophotometry (MALDI- TOF) (Microflex, Bruker Daltonics, Bremen, Germany) (Wieser et al. 2012). Each E. coli was checked for the presence of an ESBL production by using the double disk diffusion method according to the Clinical Laboratory Standard Institute guidelines (Clinical and Laboratory Standards Institute (CLSI) 2018). Escherichia coli (ATCC 25922) was used as standard strain. Evaluation of the susceptibility to 15 antibiotics including ampicillin (AMP), piperacillin (PRL), ampicillin/sulbactam (SAM), cefuroxime (CXM), cefepime (FEP), aztreonam (ATM), meropenem (MEM), Imipenem (IMP), gentamicin(CN), streptomycin (S), kanamycin (K), tetracyclin (TE), ciprofloxacin (CIP), nalidixic acid (NA), sulphamethoxazole/trimethoprim (SXT), chloramphenicol (C) was performed by disk diffusion method on Müller-Hinton Agar (MHA) using a strain suspension at 0.5 Mc Farland for inoculation. The susceptibility were reported as resistant (R), sensitive (S) and intermediate resistant (I) according to the M100 Performance Standards for Antimicrobial Susceptibility Testing (Clinical and Laboratory Standards Institute (CLSI) 2018). 6.2.3 Determination of strains phylogenetic groups Genomic DNA of these isolates was extracted by using the standard boiling method. Isolates were assigned to the four main E. coli phylogenetic groups (A, B1, B2 and D) using the triplex PCR-based method (Clermont et al. 2000). The phylogenetic classification is based on the amplification of chuA (a gene required for heme transport), yjaA (a gene detected from complete genome sequence of E. coli K-12, the function of which is unknown), and the DNA fragment TSPE4.C2 (from substractive library of E. coli). The PCR was carried out in 50 µL final reaction mixture containing 1 µL of extracted DNA, 0.2 µM of each primer (Thermofisher), 200 µM of dNTPs mix (Thermofisher), 1.25 unit of Dream Taq DNA polymerase (Thermofisher) and 1X Dream Taq Green Buffer. Samples were treated at the 154 following thermal cycling conditions: initial denaturation at 95°C for 3 min, followed by 30 cycles at 95 °C for 30s, appropriate annealing temperature (Table 6-1) for 30s, 72 °C for 30 s and a final 5 min extension at 72°C. The clonality of all strains was investigated by pulse-field gel electrophoresis (PFGE) (Brechet et al. 2014). Table 6-1 Oligonucleotid primers used for the detection of phylogenetic groups, virulence factors, pathogenic islands and antibiotic resistance genes in strains isolated from urban rivers Size Tm Gene Sequence (5' → 3') (pb) (°C) Reference Phylogroup typing (triplex PCR) chuA GACGAACCAACGGTCAGGAT 279 51 Clermont et al. (2000) TGCCGCCAGTACCAAAGACA yjaA TGAAGTGTCAGGAGACGCTG 211 51 Clermont et al. (2000) ATGGAGAATGCGTTCCTCAAC TspE4.C2 GAGTAATGTCGGGGCATTCA 151 51 Clermont et al. (2000) CGCGCCAACAAAGTATTACG Virulence factors (VFs) LT1 TCTCTATGTGCATACGGAGC 322 48 Ram et al. (2007) CCATACTGATTGCCGCAAT ST 1 CTTTCCCCTCTTTTTAGTCAG 175 47 Ram et al. (2007) TAACATGGAGCACAGGCAGG stx1 ATAAATCGCCATTCGTTGACTAC 180 48 Paton and Paton (1998) AGAACGCCCACTGAGATCATC stx2 GGCACTGTCTGAAACTGCTCC 250 49 Paton and Paton (1998) TCGCCAGTTATCTGACATTCTG hlyA GCTATGGGCCTGTTCTCCTCTGC 224 55 Ram et al. (2007) ACCACTTTCTTTCTCCCGACATCC aggR GCAATCAGATTAARCAGCGATACA 426 48 Boisen et al. (2012) CATTCTTGATTGCATAAGGATCTGG ipaH CGCGACGGACAACAGAATACACTCCATC 51 Barletta et al. (2013) 108 ATGTTCAAAAGCATGCCATATCTGTG Pathogenic islands (PAIs) pic ATTCTTCTGGCTGGCATTCC 606 48 Makobe et al. (2012) CGGGATTAGAGACTATTGTTGC irp2 AAGGATTCGCTGTTACCGGAC 287 53 Makobe et al. (2012) TCGTCGGGCAGCGTTTCTTCT eae GACCCGGCACAAGCATAAGC 384 55 Makobe et al. (2012) CCACCTGCAGCAACAAGAGG iutA GGCTGGACATCATGGGAACTGG 301 56 Makobe et al. (2012) CGTCGGGAACGGGTAGAATCG tia CGGGATCCGATGAGAGCAAAACAGGCTT 756 56 Makobe et al. (2012) GGGGTACCGAAATGATAAGTTACCCC espC GCTCAACTAAATATTGATAATGTATG 453 43 Makobe et al. (2012) CCCAGCCCCAACCCTGAAAC 155 efa/lifA GAACAAAGAACATTTTCACCAGTTC 521 48 Makobe et al. (2012) CTTTCAGGTGGGGAACCCG Antibiotic resistance genes blaTEM ATCAGCAATAAACCAGC 516 55 Shah et al. (2012) CCCCGAAGAACGTTTTC blaSHV ATGCGTTATATTCGCCTGTG 865 56 Paterson et al. (2003) GTTAGCGTTGCCAGTGCTCG blaCTX-M CGCTTTGCGATGTGCAG 593 52 Bonnet et al. (2001) ACCGCGATATCGTTGGT blaCTX-M group 1 CTTCCAGAATAAGGAATC 907 50 Bonnet et al. (2001) CCGTTTCCGCTATTACAA blaCTX-M group 2 TGACTCAGAGCATTSGCCGCT 864 57 Bonnet et al. (2001) CGYGGGTTACGATTTTCGCYGC blaCTX-M group 9 TGGTGACAAAGAGARTGCAACGG 873 61 Bonnet et al. (2001) ACAGCCCYTYGGCGATGATTCT sul1 GTGACGGTGTTCGGCATTCT 779 59 Lanz et al. (2003) TCCGAGAAGGTGATTGCGCT sul2 CGGCATCGTCAACATAACCT 721 57 Lanz et al. (2003) TGTGCGGATGAAGTCAGCTC Perreten and Boerlin GAGCAAGATTTTTGGAATCG 789 55 sul3 (2003) CTAACCTAGGGCTTTGGA Integrases int1 ACGAGCGCAAGGTTTCGGT 565 54 Su et al. (2006) GAAAGGTCTGGTCATACATG int2 GTGCAACGCATTTTGCAGG 403 52 Su et al. (2006) CAACGGAGTCATGCAGATG int3 CATTTGTGTTGTGGACGGC 717 56 Su et al. (2006) GACAGATACGTGTTTGGCAA 6.2.4 Virulence factors and pathogenic islands A total of 7 virulence factors: LT1, ST1, stx1, stx2, hlyA, aggR and ipaH and 7 pathogenic islands: O-island (efa/lifA), HPI (irp2), She (pic), LEE (eae), Tia (tia), SHI-2 (iutA) and EspC (espC), were investigated. The primer sequences and reactions conditions used in the current study are given in Table 6-1. PCR was performed in a 20 µL reaction mixture containing 1 µL of extracted DNA, 0.2 µM of each primer (Thermofisher), 200 µM of dNTPs mix (Thermofisher), 0.5 unit of Dream Taq DNA polymerase (Thermofisher) and 1X Dream Taq Green Buffer. The following thermal cycling conditions were used: initial denaturation at 95°C 156 for 3 min, followed by 30 cycles at 95 °C for 30s, appropriate annealing temperature (Table 6-1) for 30s, 72 °C for 1 min and a final 5 min extension at 72°C. 6.2.5 ESBL identification and antimicrobial susceptibility testing All E. coli exhibiting an ESBL phenotype were screened for the presence of an ESBL gene. ESBLs groups were identified by polymerase chain reaction (Table 6-1). After a screening with consensus primers targeting blaCTX-M, positive isolates were specifically targeted for blaCTX-M groups. For negative strains, presence of blaSHV and blaTEM were then tested. 6.2.6 Statistical analysis The data were analysed with R version 3.4.2 “Short summer”. A χ² test was used to determine the statistical significance of the data with a significance level of 0.05. A Multivariate ANOVA with permutation (PERMANOVA) was used to compare the abundance of E. coli phylogroups according the sampling location with a significance level of 0.05. 6.3 Results and Discussion 6.3.1 Genotyping and phylogenetic grouping of isolates 150 isolates of E. coli were isolated under selective conditions (Cefotaxime 2 µg ml-1) to focus on ESBLEC population and were then confirmed at the species level by MALDI-TOF. The analysis of the PFGE patterns revealed 67 pulsotypes of an elevated clonal diversity of cefotaxime-resistant E. coli. The isolates were assigned to four major phylogroups: A and B1 including most of the commensal strains, and B2 and D, including the most virulent extraintestinal strains. In the rivers, strains belonged mostly to the group A and B1 (74%), while 5% belonged to the group B2 and 21% belonged to the group D. The distribution of phylogenetic groups in cefotaxime resistant strains was similar between isolates from sampling sites R1, R2, R4 and R5: A (47.06 – 60%) > D (17.65 – 27.77%) > B1 (14.29 – 17.65%) > B2 (0 – 5.71) (Table 6-2). The river R3 exhibited the highest prevalence of phylogroup A and B1 resistant strains (87.5%) among the sampled sites. These results were confirmed by PERMANOVA analysis, which showed that the composition of resistant E. coli varied significantly between rivers and sampling sites (p<0.05). 157 Table 6-2 Prevalence of phylogroups according to the sampling site Phylogenetic R1 R2 R3 R4 R5 group (n=35) (n=36) (n=24) (n=17) (n=35) A 19 20 10 8 21 (54,29%) (55,56%) (41,67%) (47,06%) (60%) B1 6 6 11 3 5 (17,14%) (16,67%) (45,83%) (17,65%) (14,29%) B2 2 0 0 3 2 (5,71%) (0%) (0%) (17,65%) (5,71%) D 8 10 3 3 7 (22,86%) (27,77%) (12,5%) (17,65%) (20%) 6.3.2 Phenotypic resistance On the 150 E. coli, 124 isolates (83%) were ESBL according to the double disk diffusion method. The susceptibility of these isolates to 15 antibiotics from 11 different classes is shown Figure 6-2. Resistance or intermediate resistance from 6 to 13 antibiotics was observed among ESBLEC isolates and 97.7%, 97.2% and 99.9% of ESBLEC - isolated upstream, at the hospital effluent reject point in the river and downstream the hospital effluent - were also resistant to the 4th generation of cephalosporins. However, all isolates were susceptible to meropenem and imipenem. This finding is in agreement with existing literature as reviewed by Tadesse et al. (2017), who explained the absence of carbapenem resistance by the low availability of such drugs in Africa. Considering other classes of antibiotics than β-lactams, theses isolates were also highly resistant to tetracyclin, sulfamethoxazole/trimethoprim, nalidixic acid and to a less extend ciprofloxacin and streptomycin. This co-resistance to other antibiotic classes and particularly to quinolone, aminoglycosides and sulfonamides is due to the carriage of these resistance genes on the same type of conjugative plasmids (Wang et al. 2013, Tacao et al. 2014). Interestingly, the distribution on antibiotic resistance according to the sampling location did not show significant differences except for Gentamicin (CN) and Streptomycin (S) (χ² test, p<0.05). When considering the resistance to antibiotic classes (according Magiorakos et al. (2012) classification for Enterobacteriaceae), ESBLEC isolates showed a resistance from 5 to 10 antibiotic classes, classifiying all of them as multidrug resistant bacteria (Figure 6-3). 3.2%, 13.7%, 21.0%, 18.6%, 24.2% and 19.4% of ESBLEC isolates were resistant to 5 classes, 6 classes, 7 classes, 8 classes , 9 classes and 10 classes of antibiotics respectively and no significant difference between sampling sites nor rivers were observed (χ²-test, 0.14 158 These results are in accordance with our previous paper indicating that hospital effluents are one of several sources of antibiotic resistance dissemination in under developing countries aquatic systems (Laffite et al. 2016). Indeed, bacteria isolated upstream hospital effluent can harbour a stronger pattern of antimicrobial resistance than bacteria isolated from downstream the hospital effluent. The presence of resistant bacteria in aquatic system is nowadays a common phenomenon due to the high dissemination of ARG in the environment and resistant bacteria in river (Kittinger et al. 2016, Dhawde et al. 2018, Proia et al. 2018). In the study of Deredjian et al. (2016), the author distinguish clinical antibiotic resistance from environmental one with the resistance cut-off value of 12 antibiotics. In our study, 19 ESBLEC isolates (15.3%) were resistant to 12 or more antibiotics (i.e. 2 isolates upstream, 6 isolates at the effluent and 11 isolates downstream the hospital effluent). These findings revealed the potential contribution of hospital wastewater to the development and the dissemination of MDR isolates in rivers. 100% 90% 80% 70% 60% 50% 40% 30% Percentage Percentage of Resistance (%) 20% 10% 0% U E D U E D U E D U E D U E D U E D U E D U E D U E D U E D U E D U E D U E D U E D U E D AMP SAM CXM FEP ATM MEM IMP CN S K TE CIP NA SXT C Sampling Zone S I R na Figure 6-2 Percentages of 150 ESBL-producing E. coli isolates exhibiting antimicrobial resistance. 159 100 90 Antibiotic 80 classes 70 10 60 9 50 8 40 7 30 6 20 5 Pourcentage resistance of (%) 10 0 Upstream Effluent Downstream Figure 6-3 Prevalence of resistance to antibiotic classes in ESBLEC isolates. ESBLEC = 150 6.3.3 Distribution of antimicrobial resistance genes and mobile genetic elements in ESBLEC isolates Among the isolates, between 95.16% of ESBLEC phenotype was explained by the carriage of CTX-M gene (Figure 6-4) with a large majority (82.3%) of CTX-M group 1 genes. The CTX-M enzyme is the most predominant ESBLs among E. coli isolated from human clinical samples, and food of animal origin thus leading to a widespread dissemination of this enzyme to adjacent aquatic systems (Kürekci et al. 2017). At the reject point of hospital effluent and downstream this effluent, the presence of other ESBLEC phenotypes is generally explained by the presence of other CTX-M group, whereas upstream the effluent discharge point, ESBLEC phenotype could be explained by the presence of other ESBL enzyme such as TEM or SHV. A high prevalence of the CTX-M group 1 was detected in numerous studies (Overdevest et al. 2012, Brechet et al. 2014, Blaak et al. 2015); however the prevalence of CTX- M group 1 in our study seems to be particularly high. 160 100 90 80 70 No CTX-M 60 Other CTX-M group 50 CTX-M group 9 % of % ARGs 40 CTX-M group 3 30 CTX-M group 1 20 10 0 Downstream Effluent Upstream Figure 6-4 Comparison of the distribution (%) of extended-spectrum β-lactamases (ESBLs) between sampling location Trimethroprim-sulphamethoxazole (SXT) is a combination commonly used for the treatment of urinary tract infection (Perreten and Boerlin 2003). In the studied rivers, 71.8% of the ESBLEC isolates were resistant to this combination of antibiotics and 62.9% carried a sulphonamide resistance which was associated with the three known genes: sul1, sul2 and sul3. Among the isolates, 11.3%, 37.1% and 2.4% carried genes sul1, sul2 and sul3 alone and 12.1% of the ESBLEC carried both sul1 and sul2, respectively. To our knowledge, it is the first description of cultivable bacteria carrying sul3 resistance genes in Central Africa. All the ESBLEC isolates carried an integrase with 95.2% and 4.8% of int1 and int1-int2 carriage respectively, however no int3 was found. Class 1 integron is often use as a proxy to evaluate anthropogenic impact as it is commonly linked to genes conferring antimicrobial resistance and its abundance can rapidly evolved according to environmental selection pressure (Gillings et al. 2015). The detection of integrons is variable among studies and class I integrons was found to range between 41-64% (Kürekci et al. 2017), however the presence of Class I integrons in 100% of our isolated strains is elevated and highlight the particularly high potential of gene dissemination due to anthropogenic purposes in the studied rivers. 161 6.3.4 Association of virulence genes, DEC pathogenic islands and antibiotic resistance pattern among phylogroups Pathogenic E. coli strains differ from commensal one by the presence of VFs , which alleviate the potential of colonization and invasion of the host, avoid or disrupt the host defence mechanism or stimulate the inflammatory response (Johnson and Stell 2000). ESBLEC strains were analyzed for major VFs genes linked to diarrheagenic E. coli (DEC) and for the presence of DEC pathogenicity islands which determine the virulence potential of the bacteria. The results revealed that 15.3% (n=19/124) of the strains carried one or more VFs (Table 6-3). Among the tested VFs, only eaeA, LT1, ST1 and aggR were amplified in the ESBLEC strains. Interestingly, the distribution of VFs and PAIs does not vary significantly with the sampling zone nor the studied site (p>0.05). These results suggest that hospital effluent had no significant effect on the dissemination of DEC VFs in aquatic systems and that another source should be responsible of their dissemination. High abundance of virulence factors in environmental strains have been highlighted previously (El-Shaer et al. 2018), thus revealing the potential virulence of environmental strains. Since the pathogenicity of E. coli is linked to PAI markers, the presence of multiple PAI markers is important to estimate the virulence potential. Only, 7 ESBL strains carrying one or more VFs carried PAIs (4 strains carried one PAI and 3 strains carried 2 PAIs) with the highest prevalence in the phylogroup B2. 28.2% (n=36/124) of the strains carried one more PAIs. The high pathogenicity island (HPI irp2) and the PAI SHI-2 (iutA), which encode for a virulence factor contributing to bacterial growth in iron-limited conditions where the most prevalent. These results are in agreement with the findings of Naderi et al. (2016) and El-Shaer et al. (2018) who hypothesized that HPI is more a “fitness” than a “pathogenic” island in commensal strains. 162 Table 6-3 Phylogroup typing, antibiotic resistance, virulence factor and pathogenic island pattern of DEC strains according to their sampling location VFs PAIs ST Zone tia Typage Resistance pattern pic LT1 irp2 stx1 stx2 iutA hlyA ipaH eaeA aggR espC efa/lifA A Upstream AMP, SAM, CXM, FEP, ATM, S, TE, CIP, NA, SXT, C AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT AMP, SAM, CXM, FEP, ATM, K, TE, CIP, NA, SXT AMP, CXM, FEP, ATM, S, CIP, C AMP, SAM, CXM, FEP, ATM, TE, CIP, NA, SXT AMP, SAM, CXM, FEP, ATM, CN, TE, CIP, NA, SXT AMP, CXM, FEP, ATM, CN, CIP, NA, SXT, C Effluent AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT, C AMP, CXM, FEP, ATM, K, TE, CIP, NA, SXT AMP, SAM(*), CXM(*), FEP, ATM, CN, S, K, TE, CIP, NA, SXT, C AMP, CXM, FEP, ATM, K, TE, CIP, NA, SXT AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT, C AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT Downstream AMP, SAM, CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT, C AMP, CXM, S, TE, CIP, NA, SXT, C AMP, SAM, CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT, C AMP, SAM(*), CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT, C AMP, CXM, FEP, ATM, S, CIP, NA, SXT, C AMP, CXM, FEP, ATM, S, K, CIP, C B1 Upstream AMP, SAM, CXM, FEP, ATM, S, K, CIP, SXT AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT AMP, CXM, FEP, ATM, CN, S, CIP AMP, CXM, FEP, ATM, CN, CIP, NA Downstream AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT, C AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT, C AMP, SAM, CXM, FEP, ATM, S, TE, CIP, NA, SXT, C B2 Upstream AMP, SAM, CXM, FEP, ATM, S, K, TE, CIP, NA, SXT, C AMP, SAM, CXM, FEP, ATM, S, TE, CIP, NA, SXT AMP, CXM, FEP, ATM, S, TE, CIP, NA, SXT, C Effluent AMP, SAM(*), CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT Downstream AMP, CXM, FEP, ATM, CN, S, TE, CIP, NA, SXT AMP, CXM, FEP, ATM, CN, S, K, TE, NA, SXT AMP, CXM, ATM, S, K, TE, CIP, SXT AMP, SAM, CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT AMP, SAM, CXM, FEP, ATM, S,TE, CIP, NA, SXT, C D Upstream AMP, CXM, FEP, ATM, S, CIP, NA AMP, SAM, CXM, FEP, ATM, S, TE, CIP, SXT, C AMP, SAM, CXM, FEP, ATM, CN, K, TE, CIP, NA, SXT Effluent AMP, SAM, CXM, FEP, ATM, CIP, NA AMP, CXM, FEP, ATM, CN, S, CIP, NA, C AMP, SAM, CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT, C Downstream AMP, CXM, FEP, ATM, CN, TE, CIP, NA, SXT AMP, CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT, C AMP, SAM, CXM, FEP, ATM, CN, S, K, TE, CIP, NA, SXT, C AMP, SAM, CXM, ATM, S, TE, CIP, SXT, C AMP, SAM, CXM, ATM, S, TE, CIP,, SXT, C 6.4 Conclusion The results of this study indicate that urban rivers are spots of a large diversity of multidrug resistant strains carrying mobile genetic platforms. No significant difference on resistant phenotype were observed prior and after the discharge of hospital effluent revealing the endemicity of antibiotic resistance or the continuous discharge of resistant strains in urban rivers. The prevalence of mobile genetic platform in the studied strains highlights the potential of Sub-Saharan urban river to disseminate multiple drug resistance to more sensitive strains. 163 In developing countries, DEC are a major public health risk. The presence of multiple antibiotic resistance on such bacteria complicate the problematic of disease treatment. We found that DEC pathotypes could be resistant up to 10 different antibiotic, thus leading to a serious problem in term of therapeutic options in case of disease. Considering that hospital effluents are not drivers of virulence VFs and PAIs dissemination, we can conclude that DEC are not confined to hospital wastewater and that the implementation of proper food and water handling is a urgent need to limit the spread of E. coli pathotype. 164 References Barletta, F., E. H. Mercado, A. Lluque, J. Ruiz, T. G. Cleary and T. J. Ochoa (2013). "Multiplex Real-Time PCR for Detection of Campylobacter, Salmonella, and Shigella." Journal of Clinical Microbiology 51(9): 2822-2829. Beceiro, A., M. Tomás and G. Bou (2013). "Antimicrobial Resistance and Virulence: a Successful or Deleterious Association in the Bacterial World?" Clinical Microbiology Reviews 26(2): 185-230. Blaak, H., A. H. van Hoek, R. A. Hamidjaja, R. Q. van der Plaats, L. Kerkhof-de Heer, A. M. de Roda Husman and F. M. Schets (2015). "Distribution, Numbers, and Diversity of ESBL-Producing E. coli in the Poultry Farm Environment." PLoS One 10(8): e0135402. Boisen, N., F. Scheutz, D. A. Rasko, J. C. Redman, S. Persson, J. Simon, K. L. Kotloff, M. M. Levine, S. Sow, B. Tamboura, A. Toure, D. Malle, S. Panchalingam, K. A. Krogfelt and J. P. Nataro (2012). "Genomic Characterization of Enteroaggregative Escherichia coli From Children in Mali." Journal of Infectious Diseases 205(3): 431-444. Bonnet, R., C. Dutour, J. L. Sampaio, C. Chanal, D. Sirot, R. Labia, C. De Champs and J. Sirot (2001). "Novel cefotaximase (CTX-M-16) with increased catalytic efficiency due to substitution Asp-240-->Gly." Antimicrobial Agents and Chemotherapy 45(8): 2269- 2275. Brechet, C., J. Plantin, M. Sauget, M. Thouverez, D. Talon, P. Cholley, C. Guyeux, D. Hocquet and X. Bertrand (2014). "Wastewater Treatment Plants Release Large Amounts of Extended-Spectrum beta-Lactamase-Producing Escherichia coli Into the Environment." Clinical Infectious Diseases 58(12): 1658-1665. Bush, K. and G. A. Jacoby (2010). "Updated functional classification of beta-lactamases." Antimicrobial Agents and Chemotherapy 54(3): 969-976. Clermont, O., S. Bonacorsi and E. Bingen (2000). "Rapid and Simple Determination of theEscherichia coli Phylogenetic Group." Applied and Environmental Microbiology 66(10): 4555-4558. Clinical and Laboratory Standards Institute (CLSI) (2018). Performance Standards for Antimicrobial Susceptibility Testing 950 West Valley Road, Suite 2500, Wayne, Pennsylvania 19087 USA, 2018., Clinical and Laboratory Standards Institute. Deredjian, A., N. Alliot, L. Blanchard, E. Brothier, M. Anane, P. Cambier, C. Jolivet, M. N. Khelil, S. Nazaret, N. Saby, J. Thioulouse and S. Favre-Bonté (2016). "Occurrence of 165 Stenotrophomonas maltophilia in agricultural soils and antibiotic resistance properties." Research in Microbiology 167(4): 313-324. Dhawde, R., R. Macaden, D. Saranath, K. Nilgiriwala, A. Ghadge and T. Birdi (2018). "Antibiotic Resistance Characterization of Environmental E. coli Isolated from River Mula-Mutha, Pune District, India." Int J Environ Res Public Health 15(6). El-Shaer, S., S. H. Abdel-Rhman, R. Barwa and R. Hassan (2018). "Virulence Characteristics, Serotyping and Phylogenetic Typing of Clinical and Environmental Escherichia coli Isolates." Jundishapur Journal of Microbiology In Press(In Press): e82835. Ferguson, A. S., A. C. Layton, B. J. Mailloux, P. J. Culligan, D. E. Williams, A. E. Smartt, G. S. Sayler, J. Feighery, L. D. McKay, P. S. K. Knappett, E. Alexandrova, T. Arbit, M. Emch, V. Escamilla, K. M. Ahmed, M. J. Alam, P. K. Streatfield, M. Yunus and A. van Geen (2012). "Comparison of fecal indicators with pathogenic bacteria and rotavirus in groundwater." Science of the Total Environment 431: 314-322. Fondi, M. and R. Fani (2010). "The horizontal flow of the plasmid resistome: clues from inter- generic similarity networks." Environmental Microbiology 12(12): 3228-3242. Gillings, M. R. M., W. H. W. Gaze, A. A. Pruden, K. K. Smalla, J. M. J. Tiedje and Y. Y.-G. Zhu (2015). "Using the class 1 integron-integrase gene as a proxy for anthropogenic pollution." The ISME Journal 9(6): 1269-1279. Gonzales-Siles, L. and A. Sjoling (2016). "The different ecological niches of enterotoxigenic Escherichia coli." Environmental Microbiology 18(3): 741-751. Johnson, J. R. and A. L. Stell (2000). "Extended virulence genotypes of Escherichia coli strains from patients with urosepsis in relation to phylogeny and host compromise." J Infect Dis 181(1): 261-272. Kittinger, C., M. Lipp, B. Folli, A. Kirschner, R. Baumert, H. Galler, A. J. Grisold, J. Luxner, M. Weissenbacher, A. H. Farnleitner, G. Zarfel and R. Schuch (2016). "Enterobacteriaceae Isolated from the River Danube: Antibiotic Resistances, with a Focus on the Presence of ESBL and Carbapenemases." PLoS ONE 11(11): e0165820- e0165820. Kümmerer, K. (2009). "Antibiotics in the aquatic environment – A review – Part I." Chemosphere 75(4): 417-434. Kürekci, C., M. Aydin, M. Yipel, M. Katouli and A. Gündoğdu (2017). "Characterization of extended spectrum β-lactamase (ESBL)-producing Escherichia coli in Asi (Orontes) River in Turkey." Journal of Water and Health 15(5): 788-798. 166 Laffite, A., P. I. Kilunga, J. M. Kayembe, N. Devarajan, C. K. Mulaji, G. Giuliani, V. I. Slaveykova and J. Poté (2016). "Hospital Effluents Are One of Several Sources of Metal, Antibiotic Resistance Genes, and Bacterial Markers Disseminated in Sub- Saharan Urban Rivers." Frontiers in microbiology 7: 1128-1128. Lanz, R., P. Kuhnert and P. Boerlin (2003). "Antimicrobial resistance and resistance gene determinants in clinical Escherichia coli from different animal species in Switzerland." Veterinary Microbiology 91(1): 73-84. Magiorakos, A. P., A. Srinivasan, R. B. Carey, Y. Carmeli, M. E. Falagas, C. G. Giske, S. Harbarth, J. F. Hindler, G. Kahlmeter, B. Olsson-Liljequist, D. L. Paterson, L. B. Rice, J. Stelling, M. J. Struelens, A. Vatopoulos, J. T. Weber and D. L. Monnet (2012). "Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance." Clinical Microbiology and Infection 18(3): 268-281. Makobe, C. K., W. K. Sang, G. Kikuvi and S. Kariuki (2012). "Molecular characterization of virulence factors in diarrhoeagenic Escherichia coli isolates from children in Nairobi, Kenya." J Infect Dev Ctries 6(8): 598-604. Naderi, G., F. Haghi, H. Zeighami, F. Hemati and N. Masoumian (2016). "Distribution of pathogenicity island (PAI) markers and phylogenetic groups in diarrheagenic and commensal Escherichia coli from young children." Gastroenterology and Hepatology From Bed to Bench 9(4): 316-324. Nataro, J. P. and J. B. Kaper (1998). "Diarrheagenic Escherichia coli." Clinical Microbiology Reviews 11(1): 142-201. Overdevest, I. T. M. A., M. Heck, K. van der Zwaluw, I. Willemsen, J. van de Ven, C. Verhulst and J. A. J. W. Kluytmans (2012). "Comparison of SpectraCell RA typing and multilocus sequence typing for extended-spectrum-β-lactamase-producing Escherichia coli." Journal of Clinical Microbiology 50(12): 3999-4001. Park, J., J. S. Kim, S. Kim, E. Shin, K.-H. Oh, Y. Kim, C. H. Kim, M. A. Hwang, C. M. Jin, K. Na, J. Lee, E. Cho, B.-H. Kang, H.-S. Kwak, W. K. Seong and J. Kim (2018). "A waterborne outbreak of multiple diarrhoeagenic Escherichia coli infections associated with drinking water at a school camp." International Journal of Infectious Diseases 66: 45-50. Paterson, D. L., K. M. Hujer, A. M. Hujer, B. Yeiser, M. D. Bonomo, L. B. Rice and R. A. Bonomo (2003). "Extended-spectrum beta-lactamases in Klebsiella pneumoniae bloodstream isolates from seven countries: dominance and widespread prevalence of 167 SHV- and CTX-M-type beta-lactamases." Antimicrobial Agents and Chemotherapy 47(11): 3554-3560. Paton, A. W. and J. C. Paton (1998). "Detection and Characterization of Shiga ToxigenicEscherichia coli by Using Multiplex PCR Assays forstx 1, stx 2,eaeA, Enterohemorrhagic E. coli hlyA,rfb O111, andrfb O157." Journal of Clinical Microbiology 36(2): 598-602. Perreten, V. and P. Boerlin (2003). "A New Sulfonamide Resistance Gene (sul3) in Escherichia coli Is Widespread in the Pig Population of Switzerland." Antimicrobial Agents and Chemotherapy 47(3): 1169-1172. Poirel, L. L., P. P. Nordmann and R. A. R. Bonnin (2012). "Genetic support and diversity of acquired extended spectrum b-lactamases in Gram-negative rods." Infection, genetics and evolution 12(5): 883-893. Proia, L., A. Anzil, J. Subirats, C. Borrego, M. Farrè, M. Llorca, J. L. Balcázar and P. Servais (2018). "Antibiotic resistance along an urban river impacted by treated wastewaters." Science of The Total Environment 628-629: 453-466. Ram, S., P. Vajpayee and R. Shanker (2007). "Prevalence of Multi-Antimicrobial-Agent Resistant, Shiga Toxin and Enterotoxin ProducingEscherichia coliin Surface Waters of River Ganga." Environmental Science & Technology 41(21): 7383-7388. Rivas, L. (2015). "Detection and Typing Strategies for Pathogenic Escherichia coli." Shah, S. Q., D. J. Colquhoun, H. L. Nikuli and H. Sorum (2012). "Prevalence of antibiotic resistance genes in the bacterial flora of integrated fish farming environments of Pakistan and Tanzania." Environmental Science & Technology 46(16): 8672-8679. Shahrani, M., F. S. Dehkordi and H. Momtaz (2014). "Characterization of Escherichia coli virulence genes, pathotypes and antibiotic resistance properties in diarrheic calves in Iran." Biological Research 47(1): 28. Su, J., L. Shi, L. Yang, Z. Xiao, X. Li and S. Yamasaki (2006). "Analysis of integrons in clinical isolates of Escherichia coli in China during the last six years." FEMS Microbiology Letters 254(1): 75-80. Tacao, M., A. Moura, I. Henriques, M. Tacão and A. Correia (2014). "Co-resistance to different classes of antibiotics among ESBL-producers from aquatic systems." Water research 48: 100-107. 168 Tadesse, B. T., E. A. Ashley, S. Ongarello, J. Havumaki, M. Wijegoonewardena, I. J. González and S. Dittrich (2017). "Antimicrobial resistance in Africa: a systematic review." BMC Infectious Diseases 17(1): 616-616. UNICEF and WHO (2015). Progress on Sanitation and Drinking Water – 2015 update and MDG assessment. Geneva, Switzerland, World Health Organization. Verlicchi, P., A. Galletti, M. Petrovic, D. Barcelo and D. Barceló (2010). "Hospital effluents as a source of emerging pollutants: An overview of micropollutants and sustainable treatment options." Journal of hydrology 389(3-4): 416-428. Wang, J., R. Stephan, M. Karczmarczyk, Q. Yan, H. Hachler and S. Fanning (2013). "Molecular characterization of bla ESBL-harboring conjugative plasmids identified in multi-drug resistant Escherichia coli isolated from food-producing animals and healthy humans." Front Microbiol 4(188): 188. WHO. (2017). "Diarrhoeal disease." 26/02/2019, from https://www.who.int/news-room/fact- sheets/detail/diarrhoeal-disease. WHO (2017). Ending cholera; a global roadmap to 2030. Geneva: 22. WHO (2017). Guidelines for drinking-water quality: fourth edition incorporating the first addendum. Geneva, World Health Organization. WHO (2018). WHO report on surveillance of antibiotic consumption: 2016-2018 early implementation. Geneva, World Health Organization;. Wieser, A., L. Schneider, J. Jung and S. Schubert (2012) "MALDI-TOF MS in microbiological diagnostics—identification of microorganisms and beyond (mini review)." Applied Microbiology and Biotechnology 93, 965-974 DOI: 10.1007/s00253-011-3783-4. 169 CHAPTER 7 Prevalence of β-lactam and sulfonamide resistance genes in a freshwater reservoir, Lake Brêt, Switzerland A similar version of this chapter was published under the following reference: Laffite, A., D. M. M. Al Salah, V. I. Slaveykova and J. Poté (2019). "Prevalence of β-Lactam and Sulfonamide Resistance Genes in a Freshwater Reservoir, Lake Brêt, Switzerland." Exposure and Health. DOI 10.1007/s12403-019-00304-0 171 Abstract The spread of antibiotic resistance bacteria (ARB) and their resistance genes (ARGs) represents a great concern to public health worldwide. The aquatic ecosystems are considered as hot spot for horizontal gene transfer and sediments act as a reservoir of different contaminants. However, the occurrence of agricultural versus medical ARGs in Swiss freshwater reservoirs is understudied. Consequently, in this study, we aimed to quantify broad- spectrum β-lactam and sulfonamide resistance genes (blaTEM, blaSHV, blaCTX-M, blaNDM, sul1 and sul2) and the total bacterial load (16S rRNA genes) from the total DNA extracted from the surface sediments of the Lake Brêt, Switzerland using quantitative polymerase chain reaction (qPCR). Additionally, sediment physicochemical parameters including organic matter, grain size and toxic metal were analyzed. The results highlight the widespread dissemination of blaTEM, blaSHV and sul1, which were also highly correlated to bacterial biomass and organic matter content (R > 0.75, p < 0.05). The blaCTX-M and sul2 were occasionally present and positively correlate with the concentration of Cr, Mn, Fe and Ni, linking it to agricultural practices. 172 7.1 Introduction The use of antimicrobial agents in clinical and agricultural settings have revolutionized the treatment of infectious diseases and increased agricultural productivity (Alanis 2005, Carlet et al. 2011). However, there is great debate concerning the role of different sectors, including human and veterinary medicine, agriculture and aquatic environment, on the emergence and spread of antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs) (Graham et al. 2016). Beta-lactam antibiotics (β-lactams), are subcategorized into four families carrying a β- lactam cycle; penicillins, cephalosporins, carbapenems and monobactams. β-lactams inhibit the biosynthesis of peptidoglycan by competing with the natural substrate for penicillin-binding proteins. Resistance to β-lactams antibiotics is mediated by numerous genes summarized by Bush and Jacoby (2010). Because last generation of cephalosporins, carbapenems and monobactams, defined as critically important antibiotics, their use is highly restricted in veterinary medicine (Brügger 2010); blaTEM and blaSHV are the major β-lactam resistance genes linked to animal husbandry purposes. In veterinary medicine, sulfonamides are widely used to treat diarrhoea and other infectious diseases (SVS et al. 2016, OSAV 2017). Sulfonamides inhibit the dihydropterate synthase in the folic acid pathway of bacteria. Resistance to sulphonamide is mainly mediated by sul1, sul2 and sul3 genes which encode dihydropteroate synthases enzymes highly insensitive to sulfamides (Skold 2001). Although most attention was devoted to clinical settings, there is an actual growing interest to understand the origin and ecology of antibiotic resistance in the environment. Indeed, there are increasing evidences of an association between clinical and environmental ARGs (Baquero et al. 2008, Forsberg et al. 2012, Pruden et al. 2012, Berendonk et al. 2015). The possibility of a bidirectional exchange of ARB and ARGs between environment and clinic has been suggested (Di Cesare et al. 2014, Pasquaroli et al. 2014). The spread of anthropogenic pollutants, including antibiotics, antibiotic resistant bacteria and their associated resistance genes, in freshwater ecosystems is directly linked to partially or untreated wastewater discharge from industrial, agricultural irrigation and domestic effluents (Devarajan et al. 2015, Hussain et al. 2016, Devarajan et al. 2017). After being discharged in aquatic systems, ARGs and ARBs trend to accumulate in sediments which are considered as reservoirs of antibiotic resistance (Kümmerer 2004, Allen et al. 2010, Marti et al. 2014). Once stored, ARGs and ARBs can be remobilized form sediments and can return to the 173 clinical settings by three main routes: fishery, recreational activities and drinking water (Baquero et al. 2008, Wellington et al. 2013, Huijbers et al. 2015). Despite the application of national and international laws on antibiotic regulation and use (i.e. Alpine Convention of 1991, Natura 2000), perialpine rivers and lakes are usually exposed to high anthropogenic impact and a wide array of heavy metals, ARBs and ARGs has been found (Thevenon et al. 2012, Thevenon et al. 2013, Eckert et al. 2018). The actual risk of recreational activities in perialpine rivers and lakes contaminated by ARGs and ARBs is difficult to evaluate (Ashbolt et al. 2013) but some epidemiological studies already highlighted the link between recreational bathing and various infection such as urinary tract infection, E. coli O157:H7 and otitis (Eckert et al. 2018). The assessment of ARB and ARGs as well as their persistence overtime are key for developing and evaluating strategies to mitigate their propagation in aquatic environment (Devarajan et al. 2015). It is therefore critical to understand the fate of ARGs in aquatic environment, to evaluate potential risks and to propose scientifically sound management strategies towards improved protection of the environment and human health. The aim of the present research is thus to assess for the first time the impact of the catchment agricultural practices on the accumulation of ARGs in surface sediment of Lake Brêt, Switzerland. The assessment is based on a combination of a total bacterial load (16S rRNA genes) and selected ARGs: blaTEM, blaSHV, blaCTX-M, blaNDM, sul1 and sul2 determined quantitative polymerase chain reaction (qPCR). ARGs selection was based on various criteria including (Devarajan et al. 2016, Laffite et al. 2016): (i) clinically relevant genes for human health and (ii) genes conferring resistance to frequently used antibiotics by humans or animals in Switzerland. The sediment physicochemical parameters including organic matter, grain size and toxic metal concentrations were also analyzed and correlated with ARGs in order to identify the possible contaminant transport pathways (Poté et al., 2008). 7.2 Materials and Methods 7.2.1 Site description and sampling Lake Brêt is a small perialpine eutrophic lake located in Western Switzerland (Figure 7-1), about 10 km east of Lausanne and 3 km north of Lake Geneva at an elevation of 674 m 174 a.s.l. The lake has a surface area of 0.36 km² in regards to its relatively large catchment area of 23 km². Up to 1870, Lake Bret was a pond fed by precipitations, springwater and minor tributaries. After the construction of an adduction gallery for the derivation of the Grenet River in 1875, the original lake catchment increased by a factor of ten (2 km²). Finally, a 2.5 m-high dam was raised in the southern part of the lake to supply water for the Lausanne-Ouchy funicular. In 1922, the lake reach the volume of ca. 5 million m3 with a maximum deep of 20 m. A water treatment plant has been implemented since 1957 at the southern part of the lake to supply drinking water for the city of Lausanne (Lods-Crozet et al. 2009, Thevenon et al. 2013). The Brêt water treatment plant produces 13’000 L min-1 through a modern and complete treatment process including preozonation, flocculation, flotation, two-layer sans filtration, ozonation, activated carbon filtration and final chlorination (Service de l'eau Ville de Lausanne). Figure 7-1 Sampling site, Lake Brêt in Switzerland. B5-B19 sampling sites. In red: agricultural soil; in blue: center of the lake; in pink: edge of the lake; in yellow: Grenet derivation; in green: shore of the lake. Sampling took place in September 2017. The boat of the Department Forel for environmental and aquatic sciences was used to collect surface sediments (layer of 0-3 cm) in the lake using a “Ponar-type” grab sampler. For the sites with water depth less than 1 m, the sediments samples were taken manually. The selected sampling sites (Figure 7-1, Table 7-1) were; (i) in the middle of the lake (noticed B5 and B7), (ii) 3 m from the shore (noticed B8- B13), (iii) on the water line (B14-B19), and (iv) input location in the lake of an agricultural 175 Table 7-1 Physicochemical parameters of sediment and soil analyzed; including Swiss coordinates, depth or altitude, total organic matter (OM) content, the proportion of clay, silt and sand, and median grain size Sample Swiss coordinates Depth Organic matter (%) Carbonates (%) Clay (%) Silt (%) Sand (%) Median grain size (µm) Center of the lake B5 X: 548 869 .Y: 151 488 16.0 m 12.00 24.64 4.41 93.02 2.57 17.30 B7 X: 549 046 .Y: 151 843 4.3 m 9.20 26.39 1.19 58.29 40.52 16.51 Edge of the lake B8 X: 548 939 .Y: 151 886 3.4 m 8.11 14.33 1.49 51.06 47.45 58.23 B9 X: 548 902 .Y: 151 860 2.9 m 6.88 8.76 1.26 50.73 48.01 59.14 B10 X: 548 891 .Y: 151 823 3.5 m 6.80 10.49 0.86 34.70 64.44 104.30 B11 X: 548 856 .Y: 151 783 4.7 m 8.18 7.47 1.33 38.68 59.99 93.72 B12 X: 548 830 .Y: 151 725 4.4 m 5.61 8.66 0.60 37.81 61.59 151.00 B13 X: 548 690 .Y: 151 536 3.1 m 4.19 7.82 0.77 19.29 79.95 191.10 Shore B13 drain X: 548 690 .Y: 151 536 5 cm 9.26 8.30 1.26 29.36 69.38 139.00 B14 X: 548 802 .Y: 151 746 15 cm 3.86 2.10 22.47 72.86 4.67 9.81 B15 X: 548 855 .Y: 151 851 16 cm 2.93 11.36 1.05 37.21 61.74 104.00 B16 X: 548 919 .Y: 151 909 17 cm 2.91 1.88 18.95 63.32 17.73 14.73 Pra Romont stream B17 X: 549 032 .Y: 152 025 13.22 19.37 1.48 62.70 35.82 43.24 Grenet derivation B18 X: 549 040 .Y: 152 036 5.84 2.24 5.28 42.99 51.73 69.31 Agricultural drain drain X: 548 690 .Y: 151 536 5 cm 9.26 8.30 1.26 29.36 69.38 139.00 Shore soil B14 soil X: 548 802 .Y: 151 724 3.39 9.10 4.31 34.85 60.84 96.47 B15 soil X: 548 855 .Y: 151 829 8.77 18.48 2.23 48.44 49.33 61.43 B16 soil X: 548 919 .Y: 151 887 7.10 2.26 11.14 41.02 47.84 53.63 Agricultural soil B19 soil X: 549 347 .Y: 152 248 5.72 1.67 8.35 48.64 43.01 43.52 Contaminated control Vidy Bay (Geneva lake) X: 534 682. Y: 151 410 60 m 17.89 12.02 2.28 72.7 25.02 30.17 176 drain, Pra Romont stream and Grenet derivation. Contaminated sediments, which is under strong anthropogenic influence were sampled from the bay of Vidy and used as positive control (Devarajan et al. 2015). After collection, samples were immediately transported to laboratory in an icebox and stored at 4°C before further analysis. 7.2.2 Sediment physicochemical characterization The sediment particle grain size was measured using a Laser Coulter® LS-100 diffractometer (Beckman Coulter, Fullerton, Ca, USA), following 5 min ultrasonic dispersal in deionized water according to the method described by Loizeau et al. (1994). Sediment total water, total organic matter (OM) and carbonates (CaCO3) contents were determined by loss of weight after drying at 60°C, 550°C and 1000°C respectively. 7.2.3 Toxic metal analysis Before being analyzed, samples were freeze-dried and grounded into a fine powder. The metals (Cr, Mn, Fe, Ni, Cu, Zn, As, Cd and Pb) were determined by Inductively Coupled Plasma Mass Spectrometry (ICP-MS, Agilent model 7700 series) following the digestion of samples in Teflon bombs heated to 100°C in analytical grade 2M HNO3 (Poté et al. 2008, Devarajan et al. 2015). Multi-element standard solutions at different concentrations (0, 0.2, 1, 5, 20, 100 and 200 µg L-1) were used for calibration. Standard deviations of triplicate measurements were below 5%, and chemical blanks for the procedure were less than 2% of the sample signal. The concentrations were expressed in ppm (mg kg-1 dry weight). Total mercury (Hg) was determined by cold vapor atomic absorption spectrometry after thermal decomposition of the sample using an automatic solid analyzer (Direct Mercury Analyzer; DMA-80III, Mikrowellen-Systeme GmbH, Switzerland). The detection limit (3 SD blank) was 0.005 µg/g and the reproducibility better than 5%. The accuracy of the determination for Hg concentrations was estimated using MESS-3 (National research Council Canada); the repeated analyses never exceeded the published concentration ranges (0.09 ± 0.01 mg kg-1). 177 7.2.4 ARGs quantification by qPCR Total DNA from sediment and soil samples was extracted using PureLink™ Microbiome DNA Purification Kit (Life Technologies, Zug, Switzerland) according manufacture’s recommendation. DNA extraction was performed with three replicates from the same sample to compensate for heterogeneity. The concentration of extracted DNA was measured using the Quant- iTTM PicogreenTM dsDNA assay Kit (Life Technologies, Zug, Switzerland). The isolated DNA was stored at -20°C until used. The quantification of ARGs (blaTEM, blaSHV, blaCTX-M, blaNDM, sul1 and sul2), and 16S rRNA genes by qPCR was performed as previously described by Devarajan et al. (2015), Laffite et al. (2016) and Pei et al. (2006). Briefly, genes were quantified with Eco qPCR system (Illumina, Switzerland) using SensiFASTTM SYBR® Kit (Bioline, London, UK). The following cycling conditions were applied: 2 min at 95°C for polymerase activation; followed by 40 cycles of 95°C for 5s, optimal Tm for 10s and 72°C for 10s. The temperature melting profile was obtained using the following conditions; 95°C for 30s, optimal Tm for 30s, followed by 95°C for 30s. 7.2.5 Data analysis The 16S rRNA (total bacterial load) and the ARGs in the samples are expressed as “gene copy numbers” in per gram of dry sediment (copy g-1)/soil weight normalized to the DNA extraction yield. The “relative abundance” of ARGs were emphasized by the ratio (copy number of a gene)/(copy number of 16S rRNA) for each sample (Devarajan et al. 2015, Laffite et al. 2016). A statistical treatment of data; correlation matrix (Spearman) and Pairwise comparison was used to analyze the sampling area of the lake by monitoring quality variable (ARGs, metal content, median grain size and OM content). All the statistical analysis were done using the software R version 3.4.2 (R Core Team 2015). Linear mixed models were fitted to the data using the functionality of the package lme4 (Bates Douglas et al. 2015). Average numbers of gene copy numbers (16S rRNA gene or ARGs) were modeled using the location in the lake as fixed effects and technical and biological replicates as random effects. Significance of fixed effects was assessed by F-test using a significance level of 5%. Model checking was based on residual plots and normal probability plots using raw residuals. Pairwise comparison were evaluated base on adjusted p-values obtained using the single step method (Hothorn et al. 2008). Values below the limit of quantification (LOQ) were not included. 178 7.3 Results and discussion 7.3.1 Sediment physicochemical characteristics and metal content The grain size, OM and CaCO3 of the sediment and soil samples are shown in Table 7-1. The analyzed soil and sediment samples are silty-loamy to sandy-loamy, except the deepest point B5 in the Lake Brêt, which is silty. The content of OM in sediments from Lake Brêt values varied between 2.9 and 13.2%. Previous studies reported that OM in non-contaminated freshwater sediment varies from 0.1-6.0% (Poté et al. 2008, Haller et al. 2009). The sediment sampled at the deepest part of the lake (B5) and at the entrance of outlet pipe (drain and B17) showed an increase in organic matter content. According to our previous study (Thevenon et al. 2013), the accumulation of OM observed at the site B5 could be explained by the primary production in the lake. The high values observed in the sites B13 and B17 could be due to the presence of the outlet pipe from agricultural fields and the Grenet adduction (canal), respectively (Figure 7-1). The results of the toxic metal analysis are reported in Table 7-2. The concentration (in mg kg-1) of analyzed metals ranged from 12.47 to 41.62 for Cr, 15.81 to 53.12 for Ni, 7.41 to 28.90 for Cu, 29.18 to 73.49 for Zn, 0.8 to 0.39 for Cd, 8.01 to 26.30 for Pb and 0.02 to 0.13 for Hg. These concentrations were compared to natural background values from two large Swiss lakes (Arbouille et al. 1989, Birch et al. 1996). All the metal concentrations were comparable to the natural background values, which indicates that the accumulation of metals in the lake is a natural process. On the contrary, metal values measured in contaminated site (Bay of Vidy) indicate the potential impact of wastewater treatment plant (WWTP) effluents on the sediment quality as observed by Poté et al. (2008). Metal concentration measured in sediment were interpreted in comparison with the Sediment quality Guidelines (SQGs) for the protection of aquatic life (CCME 1999) to give an estimation of the potential hazard these sediment may represent. The toxic metals Cu, Zn, Cd, Pb and Hg in Lake Brêt were found to be below the SQGs. In some samples from the shore, the Cr is above SQGs, with maximum value of 41.62 mg kg-1 (Table 7-1). Consequently, the sediment contamination with toxic metals in Lake Brêt does not represent potential hazard for the aquatic life. 179 Table 7-2 Metal content (mg.kg-1 dry weight) of samples according to their sampling location analyzed by ICP-MS Sample Cr (mg/kg) Mn (mg/kg) Fe (mg/kg) Ni (mg/kg) Cu (mg/kg) Zn (mg/kg) As (mg/kg) Cd (mg/kg) Pb (mg/kg) Hg (mg/kg) Center of the lake B5 33.55 372.71 16945.44 44.89 28.89 73.49 1.62 0.20 14.86 0.13 B7 31.04 339.85 12420.38 38.46 25.39 70.01 0.88 0.39 18.21 0.11 Edge of the lake B8 31.58 266.03 12019.63 39.71 22.48 71.14 0.75 0.37 19.22 0.09 B9 21.98 226.70 10275.72 25.47 17.37 51.39 1.16 0.36 20.00 0.08 B10 29.30 334.82 16371.07 33.51 17.07 59.59 1.42 0.34 26.30 0.07 B11 25.26 209.94 11635.21 26.76 16.59 51.45 1.06 0.23 15.65 0.07 B12 21.95 234.18 10451.10 24.99 16.09 48.58 1.08 0.29 16.14 0.07 B13 14.91 236.30 9866.10 16.77 12.55 54.10 1.57 0.23 13.98 0.04 Shore B14 39.60 258.52 21599.91 50.92 18.14 59.66 1.26 0.21 20.50 0.5 B15 12.47 193.74 7916.15 15.81 7.41 29.18 0.97 0.08 8.01 0.03 B16 41.62 1552.32 24264.46 53.11 20.75 64.39 1.68 0.29 20.39 0.04 Pra Romont stream B17 34.42 773.61 15812.20 45.52 26.30 70.71 1.91 0.34 18.47 0.08 Grenet derivation B18 39.74 337.48 18742.53 53.12 17.03 54.19 0.63 0.15 17.83 0.07 Agricultural drain drain 17.73 217.46 10871.02 21.37 27.29 62.67 1.84 0.23 15.83 0.04 Shore soil B14 soil 27.95 331.03 21451.45 36.49 17.25 48.41 2.25 0.23 22.45 0.05 B15 soil 18.77 189.69 12793.77 21.60 9.76 33.10 1.46 0.11 15.13 0.02 B16 soil 18.80 248.22 10959.20 25.27 15.72 45.28 1.29 0.13 13.52 0.04 Agricultural soil B19 soil 33.02 859.15 20695.06 34.04 18.53 54.65 3.76 0.20 22.19 0.07 Contaminated control Vidy Bay (Geneva lake) 48.95 283.96 33720.53 38.60 192.04 447.46 8.36 1.88 66.07 1.30 SQGsa sediment 37.3 35.7 123 5.9 0.6 35 0.17 SQGsb soil 64 45 63 200 12 1.4 70 6.6 aSediment quality guidelines (mg kg-1), bSoil quality guidelines (mg kg-1), In bold: values above SQGs. 180 7.3.2 Quantification of bacterial population (16S rRNA) 16S rRNA genes abundances were firstly quantified to assess the general changes in bacterial level in the lake according to the depth of sampling and the presence of contamination input. Secondly, the 16S rRNA was used to avoid inconsistencies among qPCR assays for ARGs quantification due to sub-optimal efficiencies. The average gene copy numbers of the bacterial marker genes in the sediment samples are presented in Figure 7-2. The values ranged from 1.3x109 - 3.1x1010 copy numbers per g-1 dry sediment (g-1 DW). In the lake, bacterial abundances varied significantly (p < 0.05) according to the sampling sites. Total bacteria were most abundant at the deepest part of the Lake Brêt (2.12 ± 5.80)x1010 copy g-1). These values are similar to those observed in contaminated site (in the bay of Vidy, (3.10 ± 0.79)x1010 copy g-1) and the agricultural drain (1.6 ± 0.52)x109 copy g-1). However, this accumulation of bacteria at the center of the Lake Brêt can be linked to the accumulation of OM by sedimentation processes in the lake, thus increasing the nutrient availability and OM to promote higher bacterial count. Along the shore, the bacterial abundance is significantly decreasing (p < 0.05) and reach the value of (1.10 ± 0.16)x109 copy g-1 at 20 cm of the surface. These values are similar to those from the uncontaminated sediment (core G1) of Geneva Lake (Devarajan et al. 2015). In the derivation of Grenet adduction canal and after the discharge of Pra Romont stream, bacterial abundance was low. No significate difference was observed in comparison with the abundance measured along the shore of the lake (F-test, p > 0.05). This result highlight that the Grenet adduction canal, which is the main supplier of water for the lake Bret, does not transport high content of bacteria. Figure 7-2 Raw 16S rRNA copy number detected in soils and sediments (16S rRNA gene copy number g-1 DW). Error bars indicate standard deviation of measurements. 181 7.3.3 Quantification and contamination of antibiotic resistant genes The abundance of ARGs (blaTEM, blaSHV, blaCTX-M and blaNDM) are presented in Figure 7-3 and their normalization with by 16S rRNA in Figure 7-4. The genes blaTEM and blaSHV were found in all sampling sites, whereas blaCTX-M - only in the samples from the shore. The blaNDM gene was not identified in sediments from Lake Brêt. The ARGs copy number (ARGs g-1 of 4 5 4 5 DW) for blaTEM, blaSHV and blaCTX-M varied from 5.3x10 to 5.0x10 , from 1.7x10 to 2.6x10 4 and from no detected (ND) to 4.4x10 , respectively. The high abundance of blaTEM and blaSHV were observed in the center of the lake (sites B5 and B7). This could be explained by their ubiquitous presence in natural environment (Demanèche et al. 2008, Devarajan et al. 2015). Concerning the inputs in the sites Grenet Derivation, Pra Romont stream and agricultural drain, 4 5 4 4 the values ranged from 4.3x10 - 6.8x10 and 1.3x10 - 6.2x10 for blaTEM and blaSHV, respectively. Figure 7-3 Raw β-lactams resistance genes Figure 7-4 Normalized β-lactams resistance copy number detected in soils and genes copy number detected in in soils and sediments. The line in each box marks the sediments. The line in each box marks the media and boxes: 25th and 75th percentiles; media and boxes: 25th and 75th percentiles; whiskers: 5th and 95th percentiles and whiskers: 5th and 95th percentiles and outliers ± 1.5*IQR. outliers ± 1.5*IQR. 182 Except for sul2 (in the sites B5 and B7, center of the lake), the genes sul1 and sul2 were detected in all the studied sites (Figure 7-5). Values ranged from 8.0x104 to 7.1x105 and ND to 9.0x105 for sul1 and sul2 respectively. Interestingly, sul2 gene was identified only in sites B13, B16, B17, B18 and B19. These sites are probably more affected by the important inputs from agricultural practice. Significant difference was observed only for the sul1 in samples from shore and the other studied sites of the Lake Brêt (p<0.05). Copies numbers of ARGs quantified in lake’s Brêt input were statistically higher than those quantified in other sites (i.e. center of the lake) (p<0.05). When normalized by 16S rRNA gene copies, the great difference of genes abundance was observed (p>0.05). The values ranged from 4.6x10-3 - 5.5x10-3 sul1 copies / 16S rRNA gene and ND - 1.8x10-2 sul2 copies / 16S rRNA gene copies (Figure 7-6). Figure 7-5 Raw sulfonamides resistance Figure 7-6 Normalized sulfonamide genes copy number detected in soils and resistance genes copy number detected in in sediments. The line in each box marks the soils and sediments. The line in each box media and boxes: 25th and 75th percentiles; marks the media and boxes: 25th and 75th whiskers: 5th and 95th percentiles and percentiles; whiskers: 5th and 95th outliers ± 1.5*IQR. percentiles and outliers ± 1.5*IQR. 183 All tested genes were detected at the contaminated positive control site (Bay of Vidy), which is subjected to WWTP effluent discharge. On agricultural soil, which is only subjected to agricultural practices, blaCTX-M, blaNDM and sul1 were not quantifiable. The Lake Brêt exhibit the presence of blaTEM and blaSHV of the β-lactamases resistance genes, and sul1 and sul2 of the sulfonamides resistance genes. The ARGs abundance was found higher at the deepest part of the lake revealing an accumulation due to sedimentation processes probably of bacterial biomass and total organic matter. Such accumulation processes were already described by Zumstein (1989), who considered the deposition of natural organic matter to be due to precipitation and agricultural soil leaching. The blaTEM abundance was highly correlated with bacterial biomass and OM (R>0.75, p <0.05) and less with blaSHV (R>0.50, p<0.05). Thevenon et al. (2013) demonstrated that the primary production is a major source of OM in the Lake Brêt. The high correlation between blaTEM, bacterial biomass and OM suggest that blaTEM is disseminated in autochthonous bacteria and cannot be linked to anthropogenic input. Despite the fact that blaTEM is present in considerable quantity in both human and animal intestinal systems, this gene is known to occur in natural soil organisms (Bradford 2001, Allen et al. 2009). However, this widespread dissemination of blaSHV could be linked to the long-term use of penicillin for veterinary purposes, which remain one of the 1st line actual antibiotics with trimethoprim, sulphonamides and cephalosporin of 1st and 2nd generation in Switzerland (SVS et al. 2016). Furthermore, sul1 gene was also highly correlated to bacterial biomass (R = 0.71, p < 0.05) but less correlated to blaTEM, blaSHV and OM, thus suggesting also a widespread dissemination of sul1 gene in the lake but also the input of water enriched in sul1 genes in the lake. This result is in agreement with the long-term use of sulfonamides since 1930 and the use of sulfonamides as 1st line antibiotic for livestock in Switzerland (OSAV 2017). Contrary to β-lactamases resistance genes, sulfonamides resistance genes showed an enrichment in the surface sediment samples of the lake and in the inputs streams. This result supports our suggestion that blaTEM and blaSHV are widespread in the lake, whereas sul1 is of anthropogenic origin. Several studies demonstrated that ARGs recovered from the environment are the result of the contamination by anthropogenic sources (Wright 2010). The agricultural soil or pasture leaching, and livestock farms can be probable source of sulfonamides resistance genes found in lake Brêt. ARGs quantified in agricultural soil were 10 to 100 times smaller than those quantified in the lake sediments. Furthermore, agricultural soil was less enriched in ARGs than the other sites from the lake. Quantitative data on sulfonamides resistance genes in lakes are particularly scarce compare with those on other aquatic compartment, which make difficult 184 the comparison of the obtained results. However considering that sediment are reservoir of ARGs and taking into account the results for the sul1 abundance in the water column of 21 Swiss Lakes from Czekalski et al. (2015), it can be concluded that the Lake Brêt is fairly contaminated by sul1. Another sulfonamides resistance gene, sul2, was found only in 2 sites of the lake and in all the inputs of the lake. These results suggest that livestock should be the main driver of sul2 dissemination in the lake rather than soil leaching. Interestingly, blaCTX-M genes were only found in one sampling site and no blaNDM were detectable in the lake nor in its inputs. These results attest for the effectiveness of Swiss regulation on antibiotic use and the good st practices of antibiotics users. Indeed, blaCTX-M is linked to the use of amoxicillin which is the 1 line antibiotic for many veterinary disease (SVS et al. 2016) whereas blaNDM is linked to carpabenem use which is not allowed for veterinary practices (Brügger 2010, SwissMedic 2018). 7.3.4 ARGs and physicochemical parameter correlations The abundance of ARGs in the Lake Brêt exhibits some correlation with change of the measured sediment physicochemical parameters (Table 7-1) and the measured trace metal concentrations (Table 7-2, Figure 7-7). The genes blaTEM and blaSHV were moderately correlated (R = 0.54, p < 0.05) while the other genes showed no significant correlation (p > 0.05) between them. blaTEM and blaSHV were moderately to substantially correlated with organic matter content and carbonates (0.6>R>0.75, p<0.05). Few correlations were observed between ARGs and toxic metals. Indeed, blaTEM, blaSHV where moderately correlated with Hg (0.63 < R < 0.7, p < 0.05) and weakly correlated with Cu and Zn (0.44 < R < 0.48, p < 0.05) and blaCTX- M was found substantially positively correlated with Cr, Mn, Fe and Ni (0.63 > R > 0.85, p < 0.05) whereas sul1 was negatively correlated with Cr, Mn, Fe and Ni (-0.6 < R < -0.4, p < 0.05). Those correlations support our hypothesis of the widespread dissemination of blaTEM and blaSHV. Indeed, widespread Hg in Swiss soils is particularly linked to leaded gasoline, chimney with evacuation without filters and spreading of WWTP sludge area, whereas Cu and Zn are more linked to the long term use of these metals as feed additives (OFEV and OFAG 2008) and by their presence in mineral fertilizer recycling (Zimmermann 2018). On the other hand, Cr, Mn and Fe, were regularly found in fertilizers to overcome specific soil deficiencies (OFEV and OFAG 2012). Our results were in agreement with other studies on the correlation between ARGs and metals which exhibit weak correlation for agricultural practices and strong correlation for anthropogenic exposure (Knapp et al. 2011, Ji et al. 2012, Devarajan et al. 2015, Laffite et al. 2016). 185 The trends and correlation in ARGs and metals distribution in the lake sediments highlight the effectiveness of antibiotic regulation. Widespread ARGs such as blaTEM, blaSHV and sul1 tends to be fixed in the lake due to gene transfer processes such as HGT and their widespread dissemination in autochthonous bacteria. However, none of the more recent studied genes is well disseminated in the lake contrary to the results observed at Vidy Bay which is affected by WWTP effluent. These results suggest a low use of such antibiotics for livestock purposes due to good practices on antibiotic prescription and the absence and/or low manure spreading on agricultural soils. Concerning this assumption, we have to keep in mind that the Lake Brêt does not possess natural outflow and that the lake is known to be highly impacted by agricultural practices. So, resistance genes discharged in such aquatic system should be stored in the lake after sedimentation processes as sediment can act as gene reservoir as a mean of controlling the dissemination of these genes in the water column (Marti et al. 2014). Figure 7-7 Correlation graphic between all parameters. Color gradient represent the R coefficient. Crossmark signifies unsignificant correlation between the parameters. 186 Since the lake is a source of drinking water, the antibiotic resistance may enter in the drinking water supply system and thus, may find a pathway for human exposure to antibiotic resistant pathogens (Zhang et al. 2016). The water treatment plant play a significant role in the reduction of ARGs concentration. The cumulative effect of sand filtration, coagulation/ sedimentation and ozone sterilization, able the treatment plant to remove ARGs during the water treatment process (Zhang et al. 2016). However, some studies detected the presence of various ARGs in processed water (Schwartz et al. 2003, Hu et al. 2018). The role of the environment in the emergence and spread of ARBs and ARGs and their pathway to human health remain poorly understood, and the HGT of ARGs between indigenous environmental and pathogenic bacteria need to be quantified (Ashbolt et al. 2013). Focusing on these facts, the impact of drinking water supply system on the dissemination and maintenance of AMR need to be better understood and a monitoring of AMR in drinking water supply system and processed water is strongly recommended. 7.4 Conclusion In this study, we investigated the dissemination of ARGs in a lake used as drinking water reservoir. The results demonstrated a widespread dissemination of antibiotic-resistance genes linked to the massive use of antibiotics during many decades indicating that the lake can act as a reservoir for contaminants of emerging concern such as ARGs. The accumulation of ARGs especially, blaTEM and blaSHV in freshwater system, as found in our study, could pose a further potential threat to the humans and aquatic life. On the other hand, the Lake Brêt seems to be less-affected by broad-spectrum β-lactams and other resistance genes linked to antibiotics currently used in human and veterinary medicine. Indeed, the ban of antibiotics as growth promoters lead to a reduction of 53% of antibiotic consumption for livestock between 2008 and 2018, which should be one of the major reason for the punctual detection of blaCTX-M and sul2 in the lake. However, further research is needed to determine the temporal dissemination of bacteria and antibiotic resistance genes in the lake linked to the changes in antibiotic consumption. 187 References Alanis, A. J. (2005). "Resistance to Antibiotics: Are We in the Post-Antibiotic Era?" Archives of medical research 36(6): 697-705. Allen, H. K., J. Donato, H. H. Wang, K. A. Cloud Hansen, J. Davies and J. Handelsman (2010). "Call of the wild: antibiotic resistance genes in natural environments." Nature Reviews Microbiology 8(4): 251-259. Allen, H. K., L. A. Moe, J. Rodbumrer, A. Gaarder and J. Handelsman (2009). "Functional metagenomics reveals diverse beta-lactamases in a remote Alaskan soil." Isme Journal 3(2): 243-251. Arbouille, D., H. Howa, S. D. and J. P. Vernet (1989). Etude générale de la pollution par les métaux et répartition des nutriments dans les sédiments du Léman. Rapport Commission internationale pour la protection des eaux du Léman contre la pollution, campagne 1988. Lausanne: 139-172. Ashbolt, N. J., A. Amezquita, T. Backhaus, P. Borriello, K. K. Brandt, P. Collignon, A. Coors, R. Finley, W. H. Gaze, T. Heberer, J. R. Lawrence, D. G. Larsson, S. A. McEwen, J. J. Ryan, J. Schonfeld, P. Silley, J. R. Snape, C. Van den Eede and E. Topp (2013). "Human Health Risk Assessment (HHRA) for environmental development and transfer of antibiotic resistance." Environ Health Perspect 121(9): 993-1001. Baquero, F., J.-L. Martínez and R. Cantón (2008). "Antibiotics and antibiotic resistance in water environments." Current Opinion in Biotechnology 19(3): 260-265. Bates Douglas, Maechler Martin, W. S. Bolker Ben, Christensen Rune Haubo Bojesen, Singmann Henrik, Dai Bin and GrothendieckGabor (2015). Linear Mixed-Effects Models using 'Eigen' and S4. Berendonk, T. U., C. M. Manaia, C. Merlin, D. Fatta-Kassinos, E. Cytryn, F. Walsh, H. Bürgmann, H. Sørum, M. Norström, M.-N. Pons, N. Kreuzinger, P. Huovinen, S. Stefani, T. Schwartz, V. Kisand, F. Baquero and J. L. Martinez (2015). "Tackling antibiotic resistance: the environmental framework." Nat Rev Microbiol 13(5): 310- 317. Birch, L., K. W. Hanselmann and R. Bachofen (1996). "Heavy metal conservation in Lake Cadagno sediments: Historical records of anthropogenic emissions in a meromictic alpine lake." Water Research 30(3): 679-687. 188 Bradford, P. A. (2001). "Extended-spectrum beta-lactamases in the 21st century: characterization, epidemiology, and detection of this important resistance threat." Clin Microbiol Rev 14(4): 933-951, table of contents. Brügger, M. (2010). Directives concernant l'emploi judicieux des médicaments vétérinaires. Thörishaus, Société des Vétérinaires Suisses SVS. Bush, K. and G. A. Jacoby (2010). "Updated functional classification of beta-lactamases." Antimicrobial Agents and Chemotherapy 54(3): 969-976. Carlet, J., P. Collignon, D. Goldmann, H. Goossens, I. C. Gyssens, S. Harbarth, V. Jarlier, S. B. Levy, B. N'Doye, D. Pittet, R. Richtmann, W. H. Seto, J. W. M. van der Meer and A. Voss (2011). "Society's failure to protect a precious resource: antibiotics." The Lancet 378(9788): 369-371. CCME (1999). Recommendation canadiennes pour la qualité des sédiments. Czekalski, N., R. Sigdel, J. Birtel, B. Matthews and H. Bürgmann (2015). "Does human activity impact the natural antibiotic resistance background? Abundance of antibiotic resistance genes in 21 Swiss lakes." Environment International 81(Supplement C): 45- 55. Demanèche, S., H. Sanguin, J. Poté, E. Navarro, D. Bernillon, P. Mavingui, W. Wildi, T. M. Vogel and P. Simonet (2008). "Antibiotic-resistant soil bacteria in transgenic plant fields." Proc Natl Acad Sci U S A 105(10): 3957-3962. Devarajan, N., T. Köhler, P. Sivalingam, C. van Delden, C. K. Mulaji, P. T. Mpiana, B. W. Ibelings and J. Poté (2017). "Antibiotic resistant Pseudomonas spp. in the aquatic environment: A prevalence study under tropical and temperate climate conditions." Water Research 115: 256-265. Devarajan, N., A. Laffite, N. D. Graham, M. Meijer, K. Prabakar, J. I. Mubedi, V. Elongo, P. T. Mpiana, B. W. Ibelings, W. Wildi and J. Pote (2015). "Accumulation of clinically relevant antibiotic-resistance genes, bacterial load, and metals in freshwater lake sediments in Central Europe." Environ Sci Technol 49(11): 6528-6537. Devarajan, N., A. Laffite, C. K. Mulaji, J.-P. Otamonga, P. T. Mpiana, J. I. Mubedi, K. Prabakar, B. W. Ibelings and J. Poté (2016). "Occurrence of Antibiotic Resistance Genes and Bacterial Markers in a Tropical River Receiving Hospital and Urban Wastewaters." PLoS One 11(2): e0149211-e0149211. Di Cesare, A., S. Pasquaroli, C. Vignaroli, P. Paroncini, G. M. Luna, E. Manso and F. Biavasco (2014). "The marine environment as a reservoir of enterococci carrying resistance and 189 virulence genes strongly associated with clinical strains." Environmental Microbiology Reports 6(2): 184-190. Eckert, E. M., A. Di Cesare, M. Coci and G. Corno (2018). "Persistence of antibiotic resistance genes in large subalpine lakes: the role of anthropogenic pollution and ecological interactions." Hydrobiologia. Forsberg, K. J., A. Reyes, W. Bin, E. M. Selleck, M. O. A. Sommer and G. Dantas (2012). "The Shared Antibiotic Resistome of Soil Bacteria and Human Pathogens." Science 337(6098): 1107-1111. Graham, D. W., C. W. Knapp, B. T. Christensen, S. McCluskey and J. Dolfing (2016). "Appearance of β-lactam Resistance Genes in Agricultural Soils and Clinical Isolates over the 20th Century." Scientific Reports 6: 21550. Haller, L., E. Amedegnato, J. Pote and W. Wildi (2009). "Influence of Freshwater Sediment Characteristics on Persistence of Fecal Indicator Bacteria." Water, air and soil pollution 203(1-4): 217-227. Hothorn, T. T., F. F. Bretz and P. P. Westfall (2008). "Simultaneous Inference in General Parametric Models." Biometrical journal 50(3): 346-363. Hu, Y., L. Jiang, T. Zhang, L. Jin, Q. Han, D. Zhang, K. Lin and C. Cui (2018). "Occurrence and removal of sulfonamide antibiotics and antibiotic resistance genes in conventional and advanced drinking water treatment processes." Journal of Hazardous Materials 360: 364-372. Huijbers, P., H. Blaak, M. C. M. de Jong, E. A. M. Graat, C. M. J. E. Vandenbroucke-Grauls and A. M. de Roda Husman (2015). "Role of the Environment in the Transmission of Antimicrobial Resistance to Humans: A Review." Environmental science & technology 49(20): 11993-12004. Hussain, S., M. Naeem, M. N. Chaudhry and M. A. Iqbal (2016). "Accumulation of Residual Antibiotics in the Vegetables Irrigated by Pharmaceutical Wastewater." Exposure & health 8(1): 107-115. Ji, X., Q. Shen, F. Liu, J. Ma, G. Xu, Y. Wang and M. Wu (2012). "Antibiotic resistance gene abundances associated with antibiotics and heavy metals in animal manures and agricultural soils adjacent to feedlots in Shanghai; China." Journal of hazardous materials 235-236: 178-185. Knapp, C. W., S. M. McCluskey, B. K. Singh, C. D. Campbell, G. Hudson, D. W. Graham and J. A. Gilbert (2011). "Antibiotic Resistance Gene Abundances Correlate with Metal 190 and Geochemical Conditions in Archived Scottish Soils." PLoS ONE 6(11): e27300- e27300. Kümmerer, K. (2004). "Resistance in the environment." Journal of antimicrobial chemotherapy 54(2): 311-320. Laffite, A., P. I. Kilunga, J. M. Kayembe, N. Devarajan, C. K. Mulaji, G. Giuliani, V. I. Slaveykova and J. Poté (2016). "Hospital Effluents Are One of Several Sources of Metal, Antibiotic Resistance Genes, and Bacterial Markers Disseminated in Sub- Saharan Urban Rivers." Frontiers in microbiology 7: 1128-1128. Lods-Crozet, B., M. De La Harpe, O. Reymond and A. Stawczynski (2009). "Evaluation da la qualité chimique et biologique d’un petit lac du plateau suisse (lac de Bret, canton de Vaud)." Bulletin de la Société vaudoise des Sciences naturelles 91(4): 363-387. Loizeau, J. L., D. Arbouille, S. Santiago and J. P. Vernet (1994). "Evaluation of a wide range laser diffraction grain size analyser for use with sediments." Sedimentology 41(2): 353-361. Marti, E., E. Variatza, J. Luis Balcazar and J. L. Balcazar (2014). "The role of aquatic ecosystems as reservoirs of antibiotic resistance." Trends in Microbiology 22(1): 36- 41. OFEV and OFAG (2008). Objectifs environnementaux pour l'agriculture. A partir de bases légales existantes. Connaissance de l'environnement n°0820. Bern, Office fédéral de l'environnement: 221 p. OFEV and OFAG (2012). Elements fertilisants et utilisation des engrais dans l'agriculture. Un module de l'aide à l'exécution pour la protection de l'environnement dans l'agriculture. L'environnement pratique n°1225:63S. Berne, Office fédéral de l'environnement. OSAV (2017). ARCH-Vet: Rapport sur les ventes d'antibiotiques à usage vétérinaire en Suisse, Département fédéral de l'intérieur DFI, Office fédéral de la sécurité alimentaire et des affaires vétérinaires OSAV. 6 pages. Pasquaroli, S., A. Di Cesare, C. Vignaroli, G. Conti, B. Citterio and F. Biavasco (2014). "Erythromycin- and copper-resistant Enterococcus hirae from marine sediment and co-transfer of erm(B) and tcrB to human Enterococcus faecalis." Diagnostic microbiology and infectious disease 80(1): 26-28. Pei, R., S.-C. Kim, K. H. Carlson and A. Pruden (2006). "Effect of River Landscape on the sediment concentrations of antibiotics and corresponding antibiotic resistance genes (ARG)." Water Research 40(12): 2427-2435. 191 Poté, J., L. Haller, J.-L. Loizeau, A. Garcia Bravo, V. Sastre and W. Wildi (2008). "Effects of a sewage treatment plant outlet pipe extension on the distribution of contaminants in the sediments of the Bay of Vidy, Lake Geneva, Switzerland." Bioresource Technology 99(15): 7122-7131. Pruden, A., M. Arabi and H. N. Storteboom (2012). "Correlation Between Upstream Human Activities and Riverine Antibiotic Resistance Genes." Environmental science & technology 46(21): 11541-11549. R Core Team (2015). R: A language and environment for statistical computing. Vienne, Austria, R Foundation for Statistical Computing. Schwartz, T., W. Kohnen, B. Jansen and U. Obst (2003). "Detection of antibiotic-resistant bacteria and their resistance genes in wastewater, surface water, and drinking water biofilms." 43(3): 325-335. Service de l'eau Ville de Lausanne. Retrieved 9 novembre, 2018, from http://www.lausanne.ch/en/lausanne-officielle/administration/securite-et- economie/service-de-l-eau/derriere-le-robinet/traiter/usine-bret.html. Skold, O. (2001). "Resistance to trimethoprim and sulfonamides." Vet Res 32(3-4): 261-273. SVS, OSAV and F. Vetsuisse (2016). Stratégie d'antibiorésistance StAR. Utilisation prudente des antibiotiques: Guide thérapeutique pour les vétérinaires. SwissMedic. (2018). "Liste des substances autorisées " Retrieved 10.10.2018. Thevenon, F., T. Adatte, W. Wildi and J. Pote (2012). "Antibiotic resistant bacteria/genes dissemination in lacustrine sediments highly increased following cultural eutrophication of Lake Geneva (Switzerland)." Chemosphere 86(5): 468-476. Thevenon, F., L. F. de Alencastro, J.-L. Loizeau, T. Adatte, D. Grandjean, W. Wildi and J. Poté (2013). "A high-resolution historical sediment record of nutrients, trace elements and organochlorines (DDT and PCB) deposition in a drinking water reservoir (Lake Brêt, Switzerland) points at local and regional pollutant sources." Chemosphere 90(9): 2444-2452. Wellington, E. M. H., A. B. A. Boxall, P. Cross, E. J. Feil, W. H. Gaze, P. M. Hawkey, A. S. Johnson-Rollings, D. L. Jones, N. M. Lee, W. Otten, C. M. Thomas and A. P. Williams (2013). "The role of the natural environment in the emergence of antibiotic resistance in Gram-negative bacteria." The Lancet Infectious Diseases 13(2): 155-165. Wright, G. D. (2010). "Antibiotic resistance in the environment: a link to the clinic?" Curr Opin Microbiol 13(5): 589-594. 192 Zhang, S. S., W. W. Lin, X. X. Yu and J. J. Tiedje (2016). "Effects of full-scale advanced water treatment on antibiotic resistance genes in the Yangtze Delta area in China." FEMS microbiology, ecology 92(5): fiw065-fiw065. Zimmermann, M. (2018). Détermination des valeurs limites pour la nouvelle catégorie d'engrais "Engrais minéraux de recyclage", Département fédéral de l'économie, de la formation et de la recherche DEFR, Office fédéral de l'agriculture OFAG. Zumstein, J. (1989). Circulation des matières organiques pédogène et aquogène dans un lac eutrophe, Université de genève. 193 CHAPTER 8 Conclusions and perspectives 195 8.1 Conclusions The main objective of this study was to assess the dissemination of various groups of pollutants, including metals, POPs, fecal indicator bacteria (FIB) and antibiotic resistant genes (ARGs) in river subjected to the release of unmanaged effluents under tropical conditions as well as their potential impact on human health. Furthermore, we propose a study of the impact of the release of unmanaged effluents linked to agricultural practices and treated effluents linked to anthropogenic practices under temperate conditions in comparison of the results obtained in Africa. The major conclusions of this research are summarized below; 8.1.1 Dissemination of metals and POPs in urban rivers of Kinshasa Here, we studied the prevalence of toxic metals and POPs including OCPs, PCBs, PBDEs and PAHs in four urban rivers. In all sampling sites, high concentrations of metals and POPs were found. The study revealed that urban river in the vicinity of Kinshasa are moderately to extremely polluted by metals and can have harmful effects on aquatic fauna and flora. Furthermore, all urban rivers were also highly polluted by POPs. The origin of POPs are multiples. According to Spearman’s rank-order correlation PAHs and PBDEs originated from similar sources such as urban waste and industrial activities, combustion of coal, wood and fossil fuel whereas PCBs may originate from urban runoff. These observations highlight the heavy contamination of urban rivers by POPs and the serious threat for both human health and environment fauna and flora. 8.1.2 Surface water contamination by FIB linked to poor management of wastewaters In this study, we quantified FIB flora in an urban river and 2 shallow wells commonly used for domestic purposes. The study revealed that the physico-chemical quality of river and shallow well are overall satisfactory. However, the microbial quality of both river and shallow well were very poor. FIB quantified in river and wells exceed up to 1000 times the recommended values of US EPA. Furthermore, the study demonstrated clearly that FIB contamination originates from human activities. To finish, the study reveal a seasonal impact on the microbial quality of shallow well. During the wet season, FIB concentration increase by 2-3 orders, probably due to the higher runoff from the overflow of onsite sanitation systems. 196 The study clearly indicated the important risk for human health linked to the surface water consumption by the population. 8.1.3 Organic and inorganic pollution in urban river receiving untreated effluents In this study, we evaluate the dissemination of FIB and ARGs as well as metal pollution in the sediment of four urban rivers receiving an hospital effluent wastewater as a point source of pollution. The study reveal a high abundance of both FIB and ARGs in urban river sediment and a strong correlation between the marker genes of Enterococcus spp. and Pseudomonas with the abundance of resistance genes linking the AMR dissemination with fecal material discharge in urban rivers. Furthermore, the study showed that urban rivers are heavily contaminated by toxic metals such as Cu, Zn, Cd, Pb and Hg. The significant correlation between parameters supported the hypothesis that biological pollution could originate from common sources and that they were carried to the receiving system through common transporters. The study demonstrated that rivers were subjected to a multiple diffuse pollution and that the point-source hospital wastewater is only one of several source of AMR and metal dissemination in the aquatic system. 8.1.4 Prevalence of ESBLs and Carpapenem resistance genes in rivers receiving untreated effluents The aim of this study was to quantify the metal contamination and ARGs linked to ESBL and carbapenem resistance in urban river subjected to unregulated wastewater. Results revealed that rivers were heavily polluted by toxic metal linked to anthropogenic activities. Furthermore, genes linked to β-lactam and carbapenem resistance were detected in almost all the samples. Only the gene OXA-48 exhibit an enrichment along the river gradient meaning that β-lactam and carbapenem abundance along the river are mostly linked to the anthropogenic background of the AMR in river due to multi-diffuse fecal pollution rather from point-sources pollutions. Interestingly, despite of the low availability of carbapenem drugs in Africa, carbapenem resistance genes were detected in samples highlighting the global spread of ESBLs and CRE across the globe. 197 8.1.5 Virulence and AMR properties of ESBLs In this study, we evaluate the degree of genetic diversity among ESBLEC isolate from Kinshasa urban rivers as well as their resistance profile and virulence properties. Results showed ESBLEC had a high level of genetic diversity. Furthermore, all ESBLEC isolate were resistant to 5 to 10 classes of antibiotics, 15.3% of isolates carried one or more VFs, 28% of isolates carried one or more PAI and 100% of isolates carried an int gene. The results clearly demonstrate the strong resistance profile of ESBLEC isolate from urban river as well as their potential of dissemination by the presence of integrases and PAI. The associate of AMR and virulence properties indicate the tremendous risk associated with surface water use for domestic purposes, recreational activities and agriculture. 8.1.6 Influence of the wastewater management on the dissemination of micro- and emerging pollutants We quantified metals and ARGs in the sediments from lake Brêt (impacted by agriculture and breeding) and compared them with sediments of Vidy Bay, lake Geneva (affected by WWTP effluent waters). The results indicated that: (i) the long-term use of antibiotics fixed the ARGs linked to the oldest antibiotics used in veterinary medicine in the sediment of the lake and that they were maintained and disseminated by autochthonous organisms and; (ii) the strict Swiss regulations on antibiotic use for veterinary purposes have limited the dissemination of the newest ARGs in the lake. This finding revealed that agriculture and breeding are among ARGs drivers whereas their impact remain limited. Concerning the sediments of Vidy Bay, higher ARGs abundance, but not necessary an ARGs enrichment in comparison of the total bacterial abundance were found. This observation suggested that WWTP and thus anthropogenic influence is a primary source of ARGs abundance. The detection of MBLs identified only in Vidy Bay samples, highlighted that clinically relevant ARGs are correlated with manmade wastewaters. The same trend was observed for the dissemination of metals in aquatic systems. The results clearly indicate that the wastewater management is critical concerning the dissemination of micro- and emerging pollutants. And conventional WWTP, need to be restructured to face emerging pollutant dissemination. Actually, the WWTP is under total reconstruction (2016-2022) and a tertiary treatment process as well as a final hygenisation process will be add. 198 As described in the present study, urban environment in absence of adequate sanitation facilities represent an overall risk for human health due to the accumulation of both macro-, micro- and emerging pollutant in adjacent aquatic systems. This study revealed that the inadequate management of pollutant can be found in both developed and developing countries. Indeed, many WWTP in developed countries are not able to remove micro-pollutant efficiently before to release treated water in the environment. Data obtained in this study may help to establish baseline information on AMR dissemination in Africa. 8.2 Perspectives We propose to further pursue the research in the two studied geographical location within ongoing collaboration among University of Geneva (Switzerland) and University of Kinshasa (Congo). The following topics could be could be of particular interest for the ragion of Kinshasa in RD Congo: - Emerging pollutant such as POPs and their impact on food chains are globally understudied in developing countries settings. In the present study, we presented new data from urban location subjected to overcrowding and lack of waste management. Further studies on the impact of POPs on the local population is needed. - Performing deeper study to obtain quantitative data on ESBL and CRE from different environmental settings, and to understand different mechanisms of resistance, (co)- selection, (co)-amplification, and horizontal gene transfer (HGT) according to the climatic condition (temperate vs tropical). - Exploring the prevalence of MDR-GNB harbouring carbapenemases (KPC, NDM, OXA-48 like) in the clinical and communal wastewaters and their dissemination in to the aquatic ecosystems in developing and developed nations, and characterize their phenotypic and genotypic resistant patterns to antibiotics routinely used in prophylaxis to provide an insight on antibiotic residue, metals, antibiotic resistance genes, Mobile genetic elements (MGEs) and pathogens discharged into the aquatic ecosystems. - Collecting background information on the prevalence of antibiotics in the wastewater from hospital and communal settings to address the degree of antibiotic consumption and the role of other co-selecting agents in the dissemination of antibiotic resistance to the aquatic ecosystems. Additionally, the environmental variations such as climate (temperate vs tropical), and temperature to influence/promote the spread of Multidrug 199 Resistant Gram Negative Bacteria (MDR-GNB) in the aquatic hotspots should be performed in studied geographical location. - Characterizing potential resistance to multiple antibiotics and evaluating the link between human driven changes and environmental bacterial community, and assessing the occurrence of pathogens and MDR-GNB in raw vegetables irrigated by ground and surface water receiving systems. The role of ethnic food as a possible route for the spread of MDR-GNB from developing nations should be assessed. - Establishing correlations between climatic factors, pollutants load and the microbial communities that are the active players in the bioremediation process. In Switzerland especially in Geneva Canton, we propose the following future research topics: - To study the most prevalent and persistent CRE ST (clonal) types in hospital (e.g. HUG), communal, WWTP inlet/outlet effluents and receiving river ecosystems (Rhône River) - To identify the consortium of antibiotic degrading microorganisms from WWTP networks / contaminated sediments which may be used for bioremediation purposes - To identify the clones from WWTP, HUG and communal effluents that are capable to degrade specifically the carbapenem group of antibiotics, using functional metagenomics. 200 Curriculum Vitae 201 Amandine Laffite [email protected] Date of Birth August 21, 1986 Nationality French Mother Tongue French Other languages English (CLES level 2) Formation October 2014- PhD Student, Doctoral program in environmental sciences at the Present University of Geneva (Expected defense date: 3 June 2019) June 2014 M.S. Degree in Ecology, Microbiology, University of Lyon (France) June 2009 Bachelor Degree Biology, Ecology, University of Perpignan (France) Research experience October 2014- PhD Student, Department Forel, University of Geneva (PhD Present supervisor: Dr. John Poté ([email protected]) and PhD co- supervisor: Prof. Vera Slaveykova ([email protected])) (Expected defense date: 3 June 2019) Under the project “Evaluation of environmental impacts in water ecosystems with genetic based bacteriological and physicochemical aspects guided sustainable water resource management in developing countries: Case of south India and sub-Saharan Africa” I am evaluated the impact of wastewater management and tropical conditions on the dissemination of micropollutants such as antibiotic resistance genes and bacteria, metals and persistent organic pollutants in aquatic systems July 2014- Technician in Environmental Microbiology, UMR 5557 Microbial September 2014 Ecology, University of Lyon (France) (Research Advisor: Dr. Franck Poly and Dr. Xavier LeRoux ([email protected]) Participated in a project that aimed to determine the impact of soil management on nitrifiers and denitrifiers communities using qPCR quantitation. 203 January 2014- Master student work, UMR 5557 Microbial Ecology, University of June 2014 Lyon (France) (Research Advisor: Dr. Franck Poly and Dr. Xavier LeRoux ([email protected]) I took part in a project that aimed to determine de impact of forest species on nitrifiers communities. In the study, we used molecular biology techniques (DNA extraction and quantitation, qPCR quantitation) to determine nitrifiers abundance, phytochimic analysis (extraction of inhibitor substances from trees, HPLC analysis) and microbiological technics (bacterial growth curves). April-May 2013 Master student work, , UMR 5557 Microbial Ecology, University of Lyon (France) (Research Advisor: Dr. Zahar Haichar (feteh-el- [email protected]) I took part in a study on catabolic repression. The aim of my work was the construction of transcriptional and translational fusions for Pseudomonas brassicasearum and to test galactosidases activities Teaching experience . During the first year of my PhD studies, I co-supervised the work of Derly Yurany Barco Calderón, Master Student at the MUSE (master universitaire en sciences de l’environnement), University of Geneva. Yurani conclude her master studies presenting her work entitled “Les effets des métaux toxiques et des bactéries pathogènes provenant des eaux usées non traitées utilisées pour l’irrigation des légumes sous climat tempéré d’altitude : Le cas de Bogotá, Colombie ». I supervised the work of Abdoulaye Samake, master student ISM (Ingénieries pour la santé et le Médicament), Université de Grenoble (France). Abdoulaye ended his first year of master presenting his work entitled “Influence de la saison sur l’impact des effluents hospitaliers sur l’écosystème aquatique adjacent”. And, I supervised the work of Marie Cattin, master student at the MUSE (master universitaire en sciences de l’environnement), University of Geneva. Marie conclude her master studies presenting her work entitled “Caractérisation physico-chimique et bactériologique des sols et sediments”. . During the 2nd year of my PhD studies, I supervised the work of Joséphine Christelle Inihagbe Ngwanza, Master Student at the MUSE. Christelle conclude her master studies presenting her work entitled “ Caractérisation physico-chimique et des genes de résistance aux antibiotiques dans les sediments du lac de Brêt”. . During my 3rd year of PhD studies, I supervised the work of Carmen Diz Salgado, Master Student at the MUSE. Carmen conclude her master studies presenting her work entitled “Caractérisation physicochimique et des bactéries de résistance aux antibiotiques dans la colonne d’eau et substrat sédimentaire de certaines plages Genevoises, Lac Léman”. 204 . During my 4th year of PhD, I am co-supervising the work of Roxana O. Morales Vidal, Master Student at the master Biologie, University of Geneva. Roxana is working on her master study entitled “Quantification des gènes de résistance aux antibiotiques dans les plages du lac Léman, canton de Genève”. And, I am also co-supervising the work of Karima Shamuratova, Master Student at the MUSE. Karima is working on her master study entitled “ Prévalence et quantification des bactéries et gènes de résistance aux antibiotiques en différentes conditions de gestion des effluents urbains. ”. . Teaching assistant (years 2015, 2016, 2017) in the practical courses on ecotoxicology and water analysis for the master MUSE (Master universitaire en sciences de l’environnement) and geobotany for the Bachelor earth Sciences, University of Geneva. . PhD student coordination. During my PhD studies, I took part on the coordination of PhD students at the University of Kinshasa and the National Pedagogic University (Republic Democratic of the Congo) List of publications Laffite, A., D. M. M. Al Salah, V. I. Slaveykova and J. Poté (2019). "Prevalence of β-Lactam and Sulfonamide Resistance Genes in a Freshwater Reservoir, Lake Brêt, Switzerland." Exposure and Health. Laffite, A., P. I. Kilunga, J. M. Kayembe, N. Devarajan, C. K. Mulaji, G. Giuliani, V. I. Slaveykova and J. Poté (2016). "Hospital Effluents Are One of Several Sources of Metal, Antibiotic Resistance Genes, and Bacterial Markers Disseminated in Sub- Saharan Urban Rivers." Frontiers in microbiology 7: 1128-1128. Atibu, E. K., N. Devarajan, A. Laffite, G. Giuliani, J. A. Salumu, R. C. Muteb, C. K. Mulaji, J.-P. Otamonga, V. Elongo, P. T. Mpiana and J. Poté (2016). "Assessment of trace metal and rare earth elements contamination in rivers around abandoned and active mine areas. The case of Lubumbashi River and Tshamilemba Canal, Katanga, Democratic Republic of the Congo." Chemie der Erde - Geochemistry 76(3): 353- 362. Devarajan, N., A. Laffite, N. D. Graham, M. Meijer, K. Prabakar, J. I. Mubedi, V. Elongo, P. T. Mpiana, B. W. Ibelings, W. Wildi and J. Pote (2015). "Accumulation of clinically relevant antibiotic-resistance genes, bacterial load, and metals in freshwater lake sediments in Central Europe." Environ Sci Technol 49(11): 6528-6537. Devarajan, N., A. Laffite, C. K. Mulaji, J.-P. Otamonga, P. T. Mpiana, J. I. Mubedi, K. Prabakar, B. W. Ibelings and J. Poté (2016). "Occurrence of Antibiotic Resistance Genes and Bacterial Markers in a Tropical River Receiving Hospital and Urban Wastewaters." PLoS One 11(2): e0149211-e0149211. 205 Devarajan, N., A. Laffite, P. Ngelikoto, V. Elongo, K. Prabakar, J. Mubedi, P. M. Piana, W. Wildi and J. Poté (2015). "Hospital and urban effluent waters as a source of accumulation of toxic metals in the sediment receiving system of the Cauvery River, Tiruchirappalli, Tamil Nadu, India." Environmental Science and Pollution Research 22(17): 12941-12950. Kapembo, M. L., D. M. M. Al Salah, F. Thevenon, A. Laffite, M. K. Bokolo, C. K. Mulaji, P. T. Mpiana and J. Poté (2019). "Prevalence of water-related diseases and groundwater (drinking-water) contamination in the suburban municipality of Mont Ngafula, Kinshasa (Democratic Republic of the Congo)." Journal of Environmental Science and Health, Part A: 1-11. Kapembo, M. L., A. Laffite, M. K. Bokolo, A. L. Mbanga, M. M. Maya-Vangua, J.-P. Otamonga, C. K. Mulaji, P. T. Mpiana, W. Wildi and J. Poté (2016). "Evaluation of Water Quality from Suburban Shallow Wells Under Tropical Conditions According to the Seasonal Variation, Bumbu, Kinshasa, Democratic Republic of the Congo." Exposure & health 8(4): 487-496. Kayembe, J. M., F. Thevenon, A. Laffite, P. Sivalingam, P. Ngelinkoto, C. K. Mulaji, J.-P. Otamonga, J. I. Mubedi and J. Poté (2018). "Corrigendum to the paper: High levels of faecal contamination in drinking groundwater and recreational water due to poor sanitation, in the sub-rural neighbourhoods of Kinshasa, Democratic Republic of the Congo by Kayembe et al., (2018)." International Journal of Hygiene and Environmental Health. Kayembe, J. M., F. Thevenon, A. Laffite, P. Sivalingam, P. Ngelinkoto, C. K. Mulaji, J.-P. Otamonga, J. I. Mubedi and J. Poté (2018). "High levels of faecal contamination in drinking groundwater and recreational water due to poor sanitation, in the sub- rural neighbourhoods of Kinshasa, Democratic Republic of the Congo." International journal of hygiene and environmental health 221(3): 400-408. Kilunga, P. I., J. M. Kayembe, A. Laffite, F. Thevenon, N. Devarajan, C. K. Mulaji, J. I. Mubedi, Z. G. Yav, J.-P. Otamonga, P. T. Mpiana and J. Poté (2016). "The impact of hospital and urban wastewaters on the bacteriological contamination of the water resources in Kinshasa, Democratic Republic of Congo." Journal of environmental science and health. Part A, Toxic/hazardous substances & environmental engineering 51(12): 1034-1042. Kilunga, P. I., P. Sivalingam, A. Laffite, D. Grandjean, C. K. Mulaji, L. F. de Alencastro, P. T. Mpiana and J. Poté (2017). "Accumulation of toxic metals and organic micro- pollutants in sediments from tropical urban rivers, Kinshasa, Democratic Republic of the Congo." Chemosphere 179: 37-48. Contribution to National and International Meetings Laffite A., Slaveykova V.I. and Poté J. 2019 Effect of agricultural landscape on the dissemination of β-lactams and sulfonamides resistance genes in the lake of Brêt, Switzerland. Swiss Microbial Ecology 2019, Switzerland from 30.01 to 01.02.2019 (oral presentation) Laffite A., Slaveykova V.I. and Poté J. 2018 ESBLECs carrying multidrug resistance and virulence factors in tropical rivers receiving hospital effluents. SETAC Europe 28th Annual Meeting, Italy from 13-17 may 2018 (oral presentation) 206 Laffite A., Slaveykova V.I. and Poté 2017 J. Hospital effluents, not an exclusive source of contaminant spread in sub-saharian urban rivers. Ecotoxicomic, France from 21-24 November 2017 (poster presentation) Laffite A., Slaveykova V.I. and Poté 2017 J. Hospital effluents, not an exclusive source of contaminant spread in sub-saharian urban rivers. SETAC Europe 27th Annual Meeting from 7-11 May 2017 (Poster presentation – Young Scientist Award for the best poster presentation) Laffite A., Slaveykova V.I. and Poté J. 2015 Effect of hospital effluents discharge on accumulation of toxic metals and antibiotic resistance genes in aquatic sub-saharian sediments Swiss Microbial Ecology 2019, Switzerland from 10-12 of September, 2015 (poster presentation) References Dr. John Poté: Department F.-A. Forel, University of Geneva ([email protected]) Prof. Vera Slaveykova: Department F.-A. Forel, University of Geneva ([email protected]) Dr. Xavier Leroux: UMR 5557 Microbial Ecology, University of Lyon (France) ([email protected]) 207 208