*3.2. Phytotoxicity Test*

The phytotoxicity test was considered as a preliminary investigation to assess plant growth to provide supporting information for further investigations performed in this study.

GI% and Inh% data for Pb-soil, 85 ± 10.8% and 14 ± 10.5%, respectively, sugges<sup>t</sup> the lack of adverse conditions for plant growth. High GI% and low Inh% values indicate reduced phytotoxicity and good quality of the contaminated soil.

The results confirmed the low values of Pb bioavailability, which did not cause significant toxic effects in the species used. Moreover, thanks to the phytotoxicity test, it was possible to verify that the high values of EC did not affect the plant germination.

#### *3.3. Effect of PGPR and EDTA on Plant Growth and Pb Phytoextraction Efficiency*

Small-scale phytoremediation experiments, performed under controlled conditions (microcosm), are essential for evaluating both the species performance and the effectiveness of treatments to be used.

Therefore, the first parameter to be analyzed is biomass, which provides information about plant growth under stress conditions.

*B. juncea* and *H. annuus* species showed no visible toxic symptoms during growth in the Pb-contaminated soil, even after applying the different treatments.

The results presented in Figure 2 indicate some significant differences in dry biomass of roots and shoots between treatments in the two species.

PGPR utilization positively affected the growth of both species, both when applied alone and in combination with EDTA.

In particular, the aerial biomass of *H. annuus* increased by 13.4% and 11.5% compared to the control when PGPRs were applied individually and combined with EDTA, respectively. In contrast, a relevant effect (36.5%) on *B. juncea* can be observed only in the combined treatment (Figure 2B). As for the roots, PGPR did not significantly influence both species' biomass production, except for the combined treatment PGPR + EDTA on *B. juncea* (Figure 2A).

**Figure 2.** (**A**) Root and (**B**) shoot dry weight (mg pot−1) of *B. juncea* and *H. annuus* grown on Pb-soil control (CT) and Pb-soil treated with, PGPR, and PGPR + EDTA. The values are the mean of three replicates, and the error bars show ± standard deviation. Values with different letters are significantly different at the 5% probability level (Tukey's test).

Under the experimental conditions investigated, and it is worth highlighting this, both *B. juncea* and *H. annuus* well tolerated the amounts of Pb released by EDTA.

The absence of adverse effects of EDTA on plant growth could be attributed to its ability to chelate Pb, which prevents binding between the metal and cellular components, neutralizing cytological impacts [47,48].

Pb concentrations in roots and shoots of *B. juncea* and *H. annuus* are presented in Table 3.

**Table 3.** Effect of PGPR and PGPR + EDTA treatment on Pb concentration in root and shoot of *B. juncea* and *H. annuus* grown on Pb-soil. The values are the mean of three replicates, and the error bars show ± standard deviation. The different letters within the same column represent different significance levels at *p* < 0.05 (Tukey's test).


As suggested by SEP, which did not reveal significant amounts of metal in an immediately bioavailable form, the lowest amounts of Pb were found in the control plants of both species. These minimal concentrations could be due to the radical exudates promoting the Pb solubilization.

The application of EDTA improved Pb uptake in both species with respect to the control. Indeed, significant increases in Pb concentration in the soil solution due to the EDTA addition, as indicated by the soil bioavailability results, resulted in substantial increases in Pb concentrations in the roots and shoots of both plants.

However, concerning *B. juncea*, the highest Pb concentrations were found in the roots and shoots of plants treated with only PGPR. The shoots of *B. juncea* treated exclusively with PGPR had approximately 24 times more Pb than control shoots. The PGPR + EDTA combined treatment mainly affected *B. juncea* roots, which showed Pb values not significantly different from plants treated with PGPR alone treatment. In contrast, the PGPR + EDTA combined treatment did not increase Pb amounts in *B. juncea* shoots.

In *H. annuus,* the highest Pb concentration was found in the roots and shoots of the plants treated with PGPR + EDTA.

The phytoextraction efficiency is the relation between the metal concentration in the plants and the biomass produced by evaluating the total accumulation (total uptake), obtained by multiplying the Pb concentration in plant tissues by the corresponding biomass

produced. The effects of the treatments on Pb total uptake in *B. juncea* and *H. annuus* are illustrated in Figure 3.

**Figure 3.** Effect of PGPR and PGPR + EDTA treatment on Pb uptake by (**A**) root and (**B**) shoot of *B. juncea* and *H. annuus* grown on Pb-soil. The values are the mean of three replicates, and the error bars show ± standard deviation. Values with different letters are significantly different at the 5% probability level (Tukey's test).

Our results indicated that treatment with PGPR, alone or in combination with EDTA, effectively improved Pb phytoextraction by *B. juncea* and *H. annuus*.

In *B. juncea* roots, the combined treatment PGPR + EDTA increased Pb accumulation by about 58 times compared to the control. Concerning *B. juncea* shoots, the highest Pb accumulation was observed in plants treated with only PGPR.

In contrast, in *H. annuus*, the highest Pb accumulation in shoots and roots was attributable exclusively to the PGPR + EDTA treatment. However, bacterial inocula alone increased the *H. annuus* root biomass to the levels of the combined treatment, without showing a significant difference.

#### *3.4. NGS Analysis Results*

The bacterial communities (Figure 4) were mainly constituted by the family Pseudomonadaceae (maximum value of 51% in the sample 1 R), followed by the families Caulobacteraceae (maximum value of 30% in the sample 6 R), Xanthomonadanceae (maxi-mum value of 18% in the sample 6 R) and Sphingomonadaceae (maximum value of 15% in the sample 3S). These bacteria are often correlated to hydrocarbon and metal contamination [49,50]. Considering the known PGPR strains [51] found in these communities and mainly related to Pb tolerance, the Pseudomonadaceae and Bacillaceae were isolated in the site and used for the inoculum. Pseudomonadaceae maintained a constant abundance in all the samples, while the abundance of Bacillaceae seemed to be affected by the inoculum, with values from 2% (both Controls) to 8% (sample 2S: PGPR *B. juncea*) and to 5–10% (respectively, in samples 5S: PGPR in *B. juncea* and 6S: EDTA + PGPR in *H. annuus*). Moreover, this family can be found only in the soil samples. Similarly, the family Rhodospirillaceae, mainly represented by the genus *Azospirillum sp.*, can be found, even if only with low relative abundance (maximum value 2%), in the samples belonging to *B. juncea*. Continuing to consider PGPR families, also known for their tolerance to Pb, there is a slight increase of the Alcaligenaceae in the sample's roots in the treatment added with the inoculum. In detail, these families increase their relative abundance from 0% to 2% in sample 3 R (i.e., PGPR in *B. juncea*) and from 0% to 1–2%, respectively in the samples 5 R (i.e., PGPR in *H. annuus*) and 6 R (i.e., EDTA + PGPR in *H. annuus*). The Enterobacteraceae represents another family of interest showing its increase in the root samples treated with the mobilizing agen<sup>t</sup> EDTA in the presence of both *B. juncea* and *H. annuus* (i.e., increase from 0% to 4% in sample 2 R: EDTA with *B. juncea* and increase from 14% to 27% in sample 6 R: EDTA with *H. annuus*).

**Figure 4.** NGS Ion Torrent analysis of all the samples divided by the *B. juncea* (i.e., samples 1–3) and *H. annuus* (i.e., samples 5–8). The numbers represent the different treatments applied (i.e., Controls: 1 and 4; PGPR: 2 and 5; EDTA + PGPR: 3 and 6), and S stands for Soils and R for Roots. Sample 2 R was not included due to the low quality of the data required. Due to the high number of results obtained, the calculations, at the family level, were carried on by setting a cut-off < 2%.

Expanding the attention to other PGPR families, not known for their Pb tolerance, it can be observed that both the families Flavobacteriaceae and Nocardioacea were mainly represented in the microcosms with *H. annuus*; the first one with higher values in the root samples, while the Nocardioacea was present almost exclusively in soil samples.
