Our results showed that the presence of HS in the culture medium influences microbial antagonism. This is in accordance with previous observations made by Visser  who found that soil HS added to selective culture media influenced the number and activities of different physiological groups of microorganisms. Also, many studies [15,31–33] showed that FA have a greater effect on cells biological activities than HA compounds. In fact, a study with radiolabelled molecules indicated that due to their smaller molecular weight, FA entered plant cells while the larger aggregates of HA remained associated with the cell walls . The observed antifungal activity of R. radiobacter slightly increased in the presence of 5 mgl−1 of DPSC-FA, however, a similar effect was observed only after the addition of 50 mgl−1 of DPSC-HA (Fig. 1). In the presence of 50 mgl−1 of fossil-HA, P. ultimum inhibition by this bacterium decreased. Our results thus support previous findings [34,35] indicating that HS of compost origin seems to be more biostimulating than those of fossil origin. Low concentrations (<50 mgl−1) of FA and HA from de-inking paper sludge compost had no effect on P. ultimum inhibition by P. aeruginosa. Low concentrations of fossil-HA also had no effect on this interaction. Bacterial antifungal activity was in general decreased or completely eliminated when the RSM medium was supplemented with 500 mgl−1 of HS of either DPSC or fossil origin (Fig. 1). This suggests that the influence of HS on microbial antagonism varies with the organism tested (R. radiobacter or P. aeruginosa) and also with the origin of the HS (DPSC or fossil), concentrations (5, 50 or 500 mgl−1) and molecular weight (i.e. DPSC-FA < DPSC-HA < fossil-HA). It is also possible that the method of extracting DPSC-HA compared to the one used for fossil-HA, may have influenced bacterial antifungal activities. Our results corroborate previous statement  that the effect of HS on growth of plants and microorganisms within different experimental contexts can vary according to all of these previously mentioned factors in addition to their structure. P. ultimum inhibition by bacteria generally tended to decrease when RSM medium was supplemented with increased concentrations of HS used as sole source of Fe. This negative effect was most pronounced with fossil-HA followed by DPSC-HA and then by DPSC-FA. This pattern is also in agreement with their Fe content (Table 2). The lower Fe content of DPSC-FA probably explains why 500 mgl−1 of these compounds did not completely cancel P. ultimum inhibition by R. radiobacter in RSM medium as observed with DPSC-HA and fossil-HA (Fig. 2). However, when inorganic Fe was added in excess to RSM medium containing HS, antifungal activity of R. radiobacter was completely cancelled while that of P. aeruginosa was similar to control. The fact that R. radiobacter could no more suppress P. ultimum growth under these conditions indicates that this bacterium inhibits the fungus mainly by being the most competitive for Fe via its siderophore production. In contrast, P. aeruginosa seems to possess more than one mechanism for limiting growth of P. ultimum. In fact, this bacterium also produces HCN (Table 3) and lytic enzymes-like chitinases  in addition of well known antifungal antibiotics .
Fe is essential for microbial growth and that is why R. radiobacter growth was the slowest in the AT minimal medium free of this element. It is also under these conditions, that R. radiobacter produced its highest level of siderophore in order to survive by capturing the remaining traces of Fe. However, when FeSO4 or FA and HA from DPSC or fossil origin were added as sole source of Fe to AT medium, bacterial growth was faster. Also, in parallel, siderophore production by R. radiobacter decreased mostly in the presence of FeSO4 but also with HS additions. These results indicate that as with FeSO4, the Fe, chelated to HS, clearly influenced R. radiobacter growth and siderophore production. Siderophores of this bacterium were thus probably capable of capturing the Fe chelated on FA and HA from DPSC or fossil origin. This also means that the stability constant of Fe complexed to R. radiobacter siderophores was probably higher than the one existing between this element and the organic compounds. In fact the stability constant of Fe complexes with siderophores are generally higher than those calculated for Fe-humate complexes [1,37]. No real differences were observed between R. radiobacter growth curves in the presence of HS equivalent to 50 mgl−1 and 1.72 μM of organic Fe or in the presence of 1.72 and 20 μM of FeSO4. However, growth of R. radiobacter in AT medium free of Fe was generally more stimulated by HS rather than by FeSO4 supplements. Effectively, by comparison, growth of R. radiobacter was enhanced much more in the presence of DPSC-FA, DPSC-HA and fossil-HA at concentrations equivalent to 1.72 μM of organic Fe than in the presence of 1.72 μM of inorganic Fe. However, siderophore production by R. radiobacter was higher in the presence of HS than with FeSO4 additions. In the presence of extra FeSO4 (20 μM), siderophore production was completely inhibited. Interestingly, these results therefore suggest that the Fe chelated to HS is not available as inorganic Fe. Thus, the biostimulating effect of HS may not be related only to its Fe content. HS have possibly enhanced R. radiobacter growth by other indirect (i.e. nutritional, CEC, etc.) or direct mechanisms (i.e. electron transfer in respiration, enhancement of membrane permeability, etc.) frequently reported in literature. However, another interesting observation can be made from these results. The availability of Fe, chelated to HS, seems to vary with their nature since after 70 h incubation approximately 80%, 60% and 40% units of siderophores were measured with fossil-HA, DPSC-HA and DPSC-FA, respectively, used at concentrations equivalent to 1.72 μM of Fe. Accordingly, this effect is possibly related to their stereochemical structure . In effect, highly aromatic HS are stereochemically complex and tend to prevent easy exchange of Fe from the humic material to the naturally occurring Fe chelators. It is therefore not surprising that R. radiobacter siderophore production was more reduced with the FA compounds (DPSC-FA) than with HA compounds (DPSC-HA and fossil-HA). Also, these results again confirm that HA from DPSC are less humified or complexed than those from fossil origin. In the presence of 50 mgl−1 of DPSC-FA, DPSC-HA and fossil-HA, about 80%, 60% and 20% siderophore units were detected. This variation in siderophore production may be easily explained by the fact that these specific quantities of DPSC-FA, DPSC-HA and fossil-HA were equivalent to 1.05, 1.72 and 11.52 μM of organic Fe, respectively. These results indicate that as the organic Fe content increases, bacterial siderophore production decreases, confirming the important influence of Fe content in HS on growth of R. radiobacter and siderophore production and also on microbial antagonism mediated by Fe competition. Soil HS are usually present in physiologically inactive forms. However, chemical, biochemical and microbiological conditions of rhizosphere differ widely from those of bulk soil , and this can lead to changes in the dynamics and structure of humified organic matter. In fact, when humic macromolecules were treated with acetic acid, small-size humic fractions were obtained and thus also stimulated specific biological properties in plants [40,41]. Therefore, in rhizosphere, soil HS or those originating from addition of compost possibly play an important role in microbial activities such as antagonism and siderophore production.