Two main zones were selected on the basis of ground water and urban drains water irrigation. GPS Garmin eTrex Vista was used to registrate GPS coordinates of study sites at Ouagadougou. The site of Tanghin was located from north 12˚23.577' to East 001˚32.274' Latitude. The coordinates of Paspanga site were from north 12˚23.137' to East 001˚30.830'.

Soil sampling

Surface (0-30cm) complex soil samples were collected from two sites (Tanghin and Paspanga) for chemical soil analysis. The complex sample was collected from different locations at the same time 6.00 am (October, 2018). All the samples were completely mixed together to form a representative surface sample for analysis. The temperature of study site was ranged from 24.0˚C to 36.5˚C and 24.2˚C to 36.6˚C respectively for Tanghin and Paspanga.

Vegetable and wastewater sampling

Vegetable samples (roots and leaves of lettuce), and Low Groundwater (wastewater) were collected from two sites (Tanghin and Paspanga) of production. The samples were homogenized to obtain representative samples and were properly labelled and transported to the laboratory in clean polythene bags for analysis.

Processing of samples

The collected samples were cleaned to remove the dust particles. A part of soil and sliced vegetable samples were dried in an oven at 105˚C for 24 h until they were brittle and crisp [28] , after drying the samples were grinded into a fine powder using a commercial blender and stored in polyethylene bags, until used for acid digestion [29]. The untreated parts of soil, vegetable and wastewaster samples were used directly for microbial analysis.

Tri-acid mixture (15 ml, 70% high purity HNO₃, 65% HClO₄ and 70% H₂SO₄; 5:1:1) was added to the beaker containing 1g dry vegetable sample [29]. The mixture was then digested at 80˚C till the transparent solution was achieved.

After cooling, the digested samples were filtered using Whatman No. 42 filter paper and the filtrate was diluted to 50 ml with deionised water. Determination of the heavy metals such as, Cd, Cr, Cu, Mn, Ni, Pb and Zn in the filtrate of vegetables and atmospheric deposits was achieved by atomic absorption spectrophotometer (Shimadzu Model 6800 with graphite furnace Model GFA 7000). Thus wastewater parameters were analyzed according to the calibration methods used by the American Public Health Association for water and wastewater Analysis [30].

Transfer factor is the ratio of the concentration of heavy metals in a plant to the concentration of heavy metals in soil. The transfer factors (TF) for each heavy metal was computed based on the method described [31] [32]. The TF value for the selected heavy metals was calculated according to the following equation:

T F = [ P s ( mg ⋅ kg − 1   dry   weight ) ] / [ S t ( mg ⋅ kg − 1   dry   weight ) ] (1)

where:

P_s: plant-metal content originating from the soil, S_t: the total heavy metal contents in the soil.

Daily intake of vegetables in adult was calculated by data obtained during the survey though a questionnaire. DIM was calculated by the following equation [33] :

D I M = [ C m e t a l × C f a c t o r × D f o o d int a k e ] / B a v e r a g e   w e i g h t (2)

C_metal: Heavy metal concentrations in plants (mg∙kg⁻¹).

C_factor: Conversion factor (CF) of 0.070 was used for the conversion of fresh vegetables to dry weight.

D_{food intake}: Daily intake of the food crops was 0.50 kg∙person⁻¹∙d⁻¹.

B_{average weight}: Body weight for the adult population was 55.0 kg.

These values were used for the calculation of HRI as well.

The HRI refers to the ratio of the daily intake of metals in the food crops to the oral reference dose (RfD) [32] [34] and was calculated using the following equation:

H R I = D I M / R f D .

An HRI > 1 for any metal in food crops indicates that the consumer population faces a health risk.

The microbial analysis of samples was appreciate by the numeration of total mesophilic aerobic on Plate Count Agar at 37˚C and the count recorded after 24 hours according to standard method [35].

Yeasts and moulds were enumerated on Sabouraud agar after inoculated and incubated at 30˚C for up to 5 days according to method NF ISO21527-1 [36]. Salmonella spp. were numerated according to ISO6579 [37] at 37˚C/24 h on Salmonella-Shigella medium. The numeration of total coliform and fecal coliform 44˚C on Violet Red Bill Agar (VRBG) is according to standard method ISO4832 [38].

E. coli was numerated on Eosine Methylene Blue by ISO4832 [38]. Method of NF ISO21528-2 [39] was used to enumerate Enterobacteriaceae. Staphylococcus aureus was detected by using ISO 6888-1 [40]. The counting was done by evaluating the colonies found in media [41]. Cetrimide medium was used to detect Pseudomonas spp. for 48 h at 30˚C [42].

The recorded data were subjected to two-way analysis of variance (ANOVA) to assess the influence of different variables on the concentrations of heavy metals in the vegetables tested.

ANOVA for each vegetable was performed separately using variables such as sites. All the statistical analyses were computed with STAT5 software version 8.

Some of the physico-chemical properties of the wastewater are shown in Table 1. No significant difference (p > 0.01) was found on pH, Temperature and Dissolved oxygen. The pH of wastewater closed to FAO values and ranged from 6.58 to 6.73 respectively for Tanghin and Paspanga. Values of pH were lower than those of Oster and Shainberg [43] ranging from 7.29 to 7.45. They have reported that this range of irrigation water is not an acceptable criterion of water quality because it tends to be buffered by the soil, and most crops can tolerate a wide pH range. Temperature ranged from 25.10 to 24.90 and dissolved oxygen ranged from 3.11 to 4.67. Conductivity and turbidity presented significant difference (p < 0.01) respectively from 433.00 to 561.00 and from 4.35 to 50.03 between the two sites.

The concentration of heavy metals in the soil samples were presented in Table 2. The levels of Cd, Cr, Cu, Pb and Zn were significantly different (p < 0.01) in soil of the two sites. The mean values ranged of Cd from 3.66 to 6.60.01; Cr, from 196.39 to 218.63; Cu, from 1.01 to 6.20; Pb, from 20.65 to 40.26; Zn, from 44.56 to 53.91 mg∙kg⁻¹ soil, respectively for Tanghin and Paspanga. No significant (p > 0.01) was found concern Mg (62.96 to 65.70 mg∙kg⁻¹ soil) and Ni (15.86 to 16.18 mg∙kg⁻¹ soil). Excepted to Cd which were higher than standards limits values (1.5 mg∙kg⁻¹), the means values of heavy metals were lower than permissible limits

Values with the different letter in the same line are significantly different at p < 0.01.

for metals (mg∙kg⁻¹) in soil [44]. Also, the results obtained in this study were lower than the values reported in the soils irrigated by Uwah et al. [28]. And yet values obtained ranges exceed the permissible limits for mineral soils in arid regions [45].

The sequence of heavy metals in the soil was found in the order of Cr > Mn > Zn > Pb > Ni > Cd > Cu. A comparison of the data with the criterion established by Kabata-Pendias and Pendias [46] for the approximate concentrations of heavy metals in plants revealed that Cr was in excessive or toxic concentrations. According to Doelman [47] , N mineralization and nitrification of soil were inhibited at high concentrations of heavy metals in mg∙kg⁻¹ around 1000 for Zn, Cu and Ni, around 100 - 500 for Pb and Cr and around 10 - 100 for Cd. The rate of mineralization of soil organic matter appears to be sensitive to metal contamination in soil [48].

At high concentrations in soil, all heavy metals have also strong toxic effects on plants which results in weak growth, yield depression, disorders in plant metabolism and reduced nutrient uptake [12] [13]. Heavy metals not only inhibit root growth but also can hamper many physiological processes and, in particularly the uptake of nutrients [49]. The rate of absorption of elements by plants depends upon the cultivated plant and the soil properties, such as the pH, cation exchange capacity (CEC) and distribution of metals in different soil fractions [50]. At an acidic pH, high Mn concentrations in cultivated soils could pose a risk of toxicity to plant. Under an acidic pH, free Mn may be the predominant form in the soil solution, making it readily available for the plants. Renella et al. [51] have reported that in Mn and Zn-polluted soils, the solubility of Mn and Zn was significantly higher in the presence of organic acids, which are typically released by plant roots, thus suggesting that plants can mobilize trace elements via their root exudates.

Concentrations of Cd, Cr, Cu, Mn, Ni, Pb and Zn in different wastewater collected from Tanghin and Paspanga were given in Table 2. The sequence of heavy metals in waste water was found in the order of Zn > Cd > Mn > Pb > Cu > Ni > Cr. Excepted Cd (2.50 to 3.30 mg∙L⁻¹) and Zn (3.60 to 3.66 mg∙L⁻¹), the samples presented significant different (p < 0.01) of concentrations of heavy metals in the samples. The levels are varied quietly such as Cr (0.51 to 2.27 mg∙L⁻¹), Cu (0.88 to 3.83 mg∙L⁻¹), Mn (0.29 to 6.60 mg∙L⁻¹), Ni (0.50 to 2.30 mg∙L⁻¹), Pb (1.37 to 3.72 mg∙L⁻¹). Excepted to Pb which was found to be within the acceptable safe limit, the results showed that all of the heavy metal concentrations exceeded the permissible limits in the wastewater samples set by FAO [52] as Cd (0.2 mg∙L⁻¹), Cr (0.1 mg∙L⁻¹), Cu (0.2 mg∙L⁻¹), Mn (0.2 mg∙L⁻¹), Ni (0.2 mg∙L⁻¹), Pb (5 mg∙L⁻¹) and Zn (2 mg∙L⁻¹). Exceeding the safe limit or having a shortage of these elements may cause diverse problems for irrigated crops according to Balkhair and Ashraf [32]. The toxic and micronutrient elements may have an inhibitory effect on plant growth when their concentrations exceed the safe limit. Using wastewater in irrigation for a long period may lead to an accumulation of heavy metals in agricultural soils and plants. Therefore, the use of wastewater in irrigation leads to contamination of the food chain [53].

Levels of the heavy metals obtained in the vegetable samples are shown in Table 2. There is no significant difference in concentration of heavy metals between samples of Tanghin and Paspanga with the exception of Mn (0.29 to 6.60 mg∙kg⁻¹), (p < 0.01). Heavy metals were ranging as the concentration of Cd (1.27 to 2.93 mg∙kg⁻¹), Cr (7.28 to 7.38 mg∙kg⁻¹), Cu (0.91 to 1.70 mg∙kg⁻¹), Ni (1.74 to 2.16 mg∙kg⁻¹), Pb (1.32 to 1.69 mg∙kg⁻¹), Zn (3.08 to 3.79 mg∙kg⁻¹). These values were agreement with previous study of Uwah et al. [28].

The heavy metal levels in this study were also higher than the WHO maximum limit permissible (ML) in vegetables. The WHO maximum limit (mg∙kg⁻¹) of some metals in vegetables are: Cd (0.02), Cr (0.1 - 0.2), Cu (0.1), Mn (0.3), Ni (0.1), Pb (0.1) and Zn (0.1) [54]. In the similar study, Shagal et al. [55] noted that Lactuca sativa grown in the same soil contained Cd, Cu, Pb, Mn and Zn in variables concentrations.

The level of heavy metals in plants depends mainly on the levels of soil contamination and plant species [28]. The results from the present study demonstrated that plants grown on wastewater-irrigated soils are contaminated with heavy metals and pose a health concern.

In heavy metals polluted soils some plant species are able to accumulate fairly large amounts of heavy metals without showing stress, which represents a potential risk for animals and human health because of transmission in the food chain [14] [56]. Many people could be at risk of adverse health effects from consuming common vegetables grown in contaminated soils [57].

The results simply explained that the heavy metals accumulated in the soils

Values with the different letter in the same line are significantly different at p < 0.01.

are transported to the vegetables through their roots by the process of absorption. The absorption and accumulation of heavy metals in plant tissues depend upon many factors according to Rupa et al. [58].

A variation in the metal concentration may be due to the variable factors like heavy metal concentration in soil wastewater used for irrigation and atmospheric deposition along with the plant’s capability to uptake and accumulate the heavy metals [58].

The results of transfer factor (TF) of the heavy metals migrated from soil to lettuce of Tanghin and Paspanga sites were presented in Table 3. The TF values obtained were ranging for Cd (0.19 to 0.89), Cr (0.030 to 0.037), Cu (0.27 to 0.91), Mn (0.004 to 0.06), Ni (0.11 to 0.13), Pb (0.04 to 0.064), Zn (0.057 to 0.085) respectively in the samples of Tanghin and Paspanga. Statistical test of significance using ANOVA and t-test revealed no significant difference (p > 0.01) between the TF of heavy metals migrate from soil to lettuce samples of the two sites. The results of this study revealed that the mean values of Cd (0.63) and Cu (0.72) were closed to those found (Cd: 0.64, Cu: 0.67) by Ibrahim et al. [57]. But the mean values of TF of Cr (0.048), Mn (0.034), Ni (0.12), Pb (0.052), Zn (0.071) were lower than those reported by the same authors (Cr: 0.43, Cu: 0.67, Mn (0.68), Ni (0.36) Pb: 0.42, Zn: 0.71). The values of transfer factor of all the heavy metals in the lettuce were less than 1. According to Uwah et al. [59] , it is easy for plants species with TF > 1 to translocate metals from roots to shoots than those which restrict metals in their roots (those with TF less than1). Khan et al. [60] on the other hand, explained that if the transfer coefficient of a metal is greater than 0.50, the plant will have a greater chance of the metal contamination by anthropogenic activities. This indicates that the levels of heavy metals in the investigated samples are low but, there is a chance for the samples of Tanghin to be contaminated with Cd (TF = 0.80) and Cu (TF = 0.91) by further anthropogenic activities. So the variations in transfer factors among lettuce samples in the

Values with the different letter in the same line are significantly different at p < 0.01.

two studied sites may be attributed to differences in the concentration of metals in soil and differences in element uptake by different vegetables [61] [62].

Transfer factor is one of the key components of human exposure to heavy metals through the food chain. Transfer factors were computed for the heavy metals to quantify the relative differences in bioavailability of metals to vegetables or to identify the efficiency of a vegetable species to accumulate a given heavy metal [59]. Others authors as Lokeshwari and Chandrappa [11] then Awode et al. [63] demonstrated that TF was based on the root uptake of the heavy metals and not the foliar absorption of atmospheric metal deposits.

The data for the evaluation of the DIM and HRI from the heavy metal-contaminated lettuce in two sites of production were presented in Table 4. The DIM values were estimated according to the average vegetable consumption for adults and compared with the recommended daily intakes [64] [65].

The results showed that the DIM and HRI values were low in the lettuce crop. The DIM (mg∙kg⁻¹∙person⁻¹∙d⁻¹) of the lettuce were ranging for Cd (0.0019 to 0.0037), Cr (0.0046 to 0.0047), Cu (0.00058 to 0.0011), Mn (0.00016 to 0.0025), Ni (0.0011 to 0.0014), Pb (0.00084 to 0.0011), Zn (0.00196 to 0.0024) respectively in the samples of the two sites. These results obtained were lower compared to DIM values found by Umah et al. [28] respectively for Cr (0.012 to 0.049), Ni (0.0027 to 0.0052), Mn (0.0012 to 0.016), Pb (0.018 to 0.033) and Cd (0.001 to 0.0031). The HRI values were lower than 1 and ranged respectively for Cd (0.0019 to 0.0037), Cr (0.0046 to 0.0047), Cu (0.00058 to 0.0011), Mn (0.00016 to 0.0025), Ni (0.0011 to 0.0014), Pb (0.00084 to 0.0011), Zn (0.00196 to 0.0024). These values were lower comparing to those of Balkhair and Ashraf [32]. HRI < 1, but it could not exclude a possible future human health risk via the intake of lettuce coming from sites of Tanghin and Paspanga.

RfD: Oral reference dose (RfD); Values with the different letter in the same line are significantly different at p < 0.01.

The present study was intended to provide some assessment on the microbiological quality of lettuce and the results were assigned in Table 5. The counts of the different microorganisms enumerated were higher than the limits values of FAO/WHO [54]. The total aerobic mesophilic count (AM) ranged with significant difference (p < 0.01) from 8.25 to 9.96 log cfu∙g⁻¹ in lettuce samples respectively from Paspanga and Tanghin. This result was higher than values of reported by Soriano et al. [66] with a range of 3.0 - 7.8 log cfu∙g⁻¹. According to Pianetti et al. [67] , aerobic mesophilic count is useful for indicating the overall microbial quality of food product: generally it does not relate to food safety hazards but acts as an indicator for food quality and shelf-life duration.

The mean counts of total and fecal coliforms were ranged with no significant difference (p > 0.01) from 5.27 to 5.77 log cfu∙g⁻¹ and 4.85 to 4.99 log cfu∙g⁻¹ from lettuce, respectively. Total coliforms were higher than those (3.3 log cfu∙g⁻¹) reported by Agbossou et al. [68]. But concentration of fecal coliforms was lower comparing to 5.35 log cfu∙g⁻¹ numerated by Thiaw [69]. E. coli were detected with no significant difference (p > 0.01) and ranged from 4.01 to 4.74 log cfu∙g⁻¹. Salmonella spp. was reported with significant difference (p < 0.01) in the analyzed samples (25 g of food) from to 2.07 to 3.12 log cfu∙g⁻¹ respectively for Paspanga and Tanghin.

The microflora of vegetables is diverse but consists predominantly of gram negative bacteria such as Enterobacteria. The range for this group in both samples was 3.92 to 5.48 log cfu∙g⁻¹ and presented significant difference (p < 0.01). Agbossou et al. [69] found lower value (2 log cfu∙g⁻¹).

The level of Enterobacteria found in this study is common in raw vegetables and are not necessarily associated with fecal contamination, because the majority of the genera are part of the endogenous microflora of the product.

Staphylococcus spp. occurred with no significant difference (p > 0.01) respectively

Values with the different letter in the same line are significantly different at p < 0.01.

2.35 to 3.10 log cfu∙g⁻¹. Cesar et al. [70] assumed that Staphylococcus contamination could cause production of toxins in lettuce when count over 5 log cfu∙g⁻¹.

Pseudomonas spp. mean counted only on lettuce samples of Tanghin (0.30 log cfu∙g⁻¹) and was lower those of Pianetti et al. [67] in ready-to-eat vegetables salads (7 log cfu∙g⁻¹). This germ is undesirable because it is often responsible for spoilage of fresh vegetables due to the production of pectinolytic enzymes which cause breakdown of the peptic polymers in plant cells [71].

Yeasts and moulds (YM) mean counts were 5.88 to 6.97 log cfu∙g⁻¹from Paspanga and Tanghin lettuce, respectively. These means were higher those of Abadias et al. [72]. Tournas and Katsoudas [73] referred the possible health problems associated with the presence of moulds in vegetables, as some may produce mycotoxins and others are known to cause allergies when they are able to produce large numbers of conidia.