The microbiological materials used were plant pathogenic fungi Botrytis cinerea , B. fabae , Colletotrichum acutatum , Fusarium oxysporum f. sp. ciceris , F. o . f. sp. lycopersici , F. o . f. sp. melonis , Verticillium dahliae , Verticillium albo-atrum and Alternaria alternata ; an oomycete ( Phytophthora cinnamomi ), one phytopathogenic bacterium ( Pseudomonas syringae ); one fungal biocontrol agent ( Trichoderma harzianum ) and three species of Rhizobium viz., Rhizobium etli , Rhizobium meliloti , Rhizobium tropici that fix atmospheric nitrogen in symbiotic association with roots of leguminous crops. Table 1 shows their collection sources and references.

All the fungal species and P. cinnamomi were retrieved and maintained in PDA medium (20 g/L dextrose, 4 g/L potato extracts, 15 g/L agar). P. syringae was maintained in Nutrient Broth/Agar I medium (5 g/L beef extract, 10 g/L peptone, 5 g/L NaCl , 15 g/L agar made up according to supplier instructions), and Rhizobium spp. Were kept in TY medium (0.5 g/L CaCl ₂ , 5 g/L tryptone, 3 g/L yeast extract, 15 g/L agar).

All the above mentioned microorganisms were tested against different doses of commercial CuNPs, provided by M.K. Impex Corp., Ontario L5N 6X1, Canada (product MKN-Cu-050, CAS number 7440-50-8, 99.9% purity, 50 nm APS, 12 m ² /g of specific surface area (SSA)). Required quantity of CuNPs was weighed and added to a sterile test tube containing 10 mL of sterile distilled water. The tube was held in an ultrasonic bath (USC Series ‘TH’; Device type USC 900; 45kHz ultrasound frequency) for 60 s to evenly disperse the CuNPs. The nanoparticle dispersion was added to sterile, melted microbial media and stirred before plating.

The commercial form (‘Ossirame 50 wg’, ManicaCobre S.L.; 50% a.i) of the compound copper oxychloride (CoC) was used as positive control and as counterpart of CuNPs for comparison. The Petri plates without any addition of CuNPs or CoC served as negative control. CuNPs and CoC were mixed at various proportions given in ‘Treatment’ section below to study their synergistic effect as well.

Initially, all the fungal materials including the oomycete P. cinnamomi were tested against either copper formulation (CuNPs and CoC) at 100 mg/L in the growth media. Subsequently, P. cinnamomi was grown on the medium containing 50 mg/L of both CuNPs and CoC, to check their synergistic effect in vitro . The plant pathogenic fungus A. alternata was further exposed to concentrations of 100, 200 and 400 mg/L of CuNPs alone and in combination with CoC (CuNP:CoC 50:50, 100:100, 200:200, and 400:400 mg/L) to test its growth and sporulation behaviour. Similarly, growth and sporulation of the biocontrol fungus T. harzianum was tested in a medium containing 10, 100 or 1000 mg/L of CuNPs.

The bacterium P. syringae was tested against 50, 100, 200 and 400 mg/L of CuNPs alone and also in combination with CoC (CuNP:CoC 50:50, 100:100, 200:200, and 400:400 mg/L) in vitro . The three Rhizobium spp. ( R. tropici , R. etli and R. meliloti ) were tested against lower concentrations of CuNPs alone (25, 50, 100, 200 mg/L) and in combination with CoC (CuNP:CoC 25:25, 50:50, 100:100 mg/L). The Petri plates without any addition of CuNPs or CoC served as negative control.

To know the effect of CuNPs on germination and other growth behaviours of wheat seeds, seeds cv. ‘Gazul’ were grown in presence of 1, 5, 10, 25, 50 and 100 mg/L of CuNPs in aqueous suspension in vitro .

Six millimetre diameter mycelial discs were cut from the leading edge of three days old fungal cultures and used for inoculation of all the fungi and P. cinnamomi in Petri dishes containing the corresponding culture media (PDA) and the different treatments.

For the plant pathogenic bacterium P. syringae , one bacterial colony of ~2 mm diameter from two-day old culture was picked and transferred to 10 mL sterile distilled water in a sterile test tube. The tube was shaken manually to obtain a bacterial suspension. The suspension thus made, was serially diluted to 10 ^-3 concentration of the original and 50 µL of the final dilution was transferred to each Petri dish prepared two days ahead with the corresponding culture medium (Nutrient Broth/Agar I) and treatments. The bacterial inoculum was spread on the medium surface with a sterile glass spreader. A similar protocol was followed for Rhizobium spp., except that the colony diameter varied based on the type of species and the culture medium (TY medium).

The cultures of all the fungi, the oomycete P. cinnamomi and all the bacteria under study were grown in a BOD incubator maintained at 25°C.

The growth of all the fungi and the oomycete P. cinnamomi was scored measuring radial mycelial growth in Petri dishes with a digital calliper, seven days after inoculation. Bacterial growth was measured by counting and recording the number of colonies appearing on the medium surface after 3-6 days, depending on the species. In addition, T. harzianum and A. alternata were further assessed for sporulation. In the case of T. harzianum , spore production was recorded every day, beginning two days after inoculation. For A. alternata , plates were kept in alternate light and dark cycles seven days after inoculation and observations on spore production were measured 7, 14 and 21 days after inoculation.

Wheat seeds of the cultivar ‘Gazul’ were used for the experiment. For each replication under various treatments 20 seeds were weighed initially. A germination test was conducted in 85 mm disposable Petri dish in the presence of aqueous suspension of CuNPs and CoC separately. Twenty seeds were placed in each plate containing 20 mL of aqueous suspension of either CuNPs or CoC, at concentrations of 1, 5, 10, 25, 50 and 100 mg/L. Sterile distilled water was used as a negative control. Observation of seed germination was made every 24-hour interval until the 5 ^th day. A seed was regarded as germinated when both radicle and plumule emerged with a length larger than 2 mm ( Sikder & Paul, 2010 ).

On the 5 ^th day, root and shoot lengths for all the seedlings were measured. Roots and shoots were excised carefully from each seedling and dried in hot air oven at 70°C for 72 hours. The remaining seeds after separating the roots and shoots were also oven-dried. Dry weight was recorded for all of them.

For all the assays, every treatment and concentration for each of the microorganisms was replicated three times. All the experiments were conducted using a randomized experimental design and repeated at least twice. For the wheat experiments, three replications (thus 60 seeds altogether in one treatment) were performed for each treatment using a randomized design. ANOVA and HSD were calculated using Statistix 8.0 for Windows.

The percent inhibition of fungal growth in vitro was calculated using the following formula ( Lamsal et al ., 2011 ):

where R =radial growth of the fungi in control plate, r = radial growth of the fungi in treatment plate.

The rate of germination was calculated according to the following formula proposed by Krishnasamy & Seshu (1990) :

CuNPs were not effective in inhibiting the mycelial growth of most of the tested fungi ( Fig. 1A ). However, significant inhibition was observed for the oomycete P. cinnamomi, against which CuNPs recorded near 60% growth inhibition compared to 46% with a commercial formulation of CoC at a dose of 100 mg/L. More interestingly, a synergistic effect of the two products could be observed resulting in inhibition of 76% when they were mixed at 50 mg/L each ( Fig. 2A ). Other plant pathogenic fungi showed variable responses for growth inhibition and growth promotion with CuNPs, CoC and their combinations. Growth reduction of the fungal pathogens B. cinerea , C. acutatum , V. dahliae and V. albo-atrum by CuNPs was observed but at low level (less than 10%) ( Fig. 1A ). Enhancement of mycelial growth of plant pathogenic fungi B. fabae , F. oxysporum f. sp. ciceris , F. oxysporum f. sp. melonis and A. alternata was recorded to various degrees by CuNPs at 100 mg/L with the highest being in A. alternata (23%). The highest degree of growth promotion (32%) was observed in F. oxysporum f. sp. ciceris in CoC treatment compared to only 7% with CuNPs ( Fig. 1A ).

Regarding A. alternata , the pathogen was subjected to further experiment with CuNPs at higher doses. It is revealed from Fig. 2B that 800 mg/L dose of CuNPs completely inhibited mycelial growth of A. alternata . Interestingly, a synergistic effect of CoC and CuNPs was observed with complete inhibition at 400 mg/L dose of each of them combined ( Fig. 1B ). The synergistic inhibitory effect was more pronounced on A. alternata in terms of sporulation. Spores were not detected in Petri dishes treated with 400 mg/L CuNPs, nor in dishes treated with 200 mg/L of both CoC and CuNPs ( Fig. 1C ). There was a significant difference in the number ofspores formed (0.6, 2.4 and 4.1) with CoC and CuNPs treatment (100 mg/L each) compared to control (10.6, 14.4, 34.3) at one, two and three weeks after inoculation respectively ( Fig. 1C ).

The biocontrol fungus T. harzianum was not inhibited at all in terms of mycelial growth ( Fig. 1A ). However, earlier and higher rate of sporulation of T. harzianum were recorded in CuNPs (100 mg/L) treated Petri dishes till 8 days after inoculation (DAI) ( Fig. 1E ), though the final sporulation count at 14 DAI scored a spore density which is not statistically different from that of control plates. CoC treatment also induced higher sporulation of the fungus till 8 DAI. The mycelial growth of T. harzianum was completely inhibited only at 1000 mg/L of CuNPs and CoC when tested separately (Figs. 1D, 2C).

Growth of P. syringae was completely inhibited in vitro with CuNPs at 200 mg/L ( Fig. 3 ). At lower doses, CoC showed more inhibitory effect than CuNPs at 100 mg/L concentration. However, significantly higher numbers of growing colonies of P. syringae were observed at 50 and 100 mg/L concentrations of CuNPs as well as CoC, compared to control plates. Six to eight-fold higher number of colonies of P. syringae was observed in growth media treated with CuNPs at concentrations lower than 200 mg/L compared to control. The synergistic effect of CuNPs and CoC was prominent when tested in the combination of 50 mg/L each. Further, combined 100 mg/L concentration of both CuNPs and CoC completely stopped P. syringae growth in vitro .

Colonies of the three Rhizobium spp., R. etli , R. meliloti and R. tropici , appeared six, five and three DAI, respectively, due to the differences in their growth traits. The three Rhizobium spp. responded differently to the CuNPs and CoC treatments. It appears from the data ( Fig. 4 ) that copperF is not inhibitory to R. tropici growth up to 200 mg/L. R. etli was inhibited completely only at 200 mg/L CoC but not in the presence of CuNPs ( Fig. 5 ). In contrast, R. meliloti was inhibited at a lower concentration of 100 mg/L CoC and 50 mg/L of either CuNPs and CoC ( Fig. 6 ), showing more sensitivity to copper compared with the other two species.

CuNPs affected wheat seed germination, root length, shoot length, root and shoot dry weight of wheat seedlings in all the concentrations. The rate of germination, co-efficient of germination and germination vigour index of the wheat seeds germinated in presence of CuNPs are presented in Fig. 7. Results showed that for the three germination traits, CuNPs significantly influenced only germination vigour index. Overall, rate of germination of wheat seeds ( Fig. 7A ) showed tendency to a higher value than control in presence of CuNPs and CoC. However, with increasing concentration of CuNP and CoC the rate of germination did not show a clear increasing or decreasing tendency nor significant differences with control. On the other hand, germination vigour index of wheat seeds was recorded to be gradually declining with increasing concentration of CuNPs and CoC ( Fig. 7C ).

Shoot length was more or less unaffected in presence of CuNPs and CoC, with no significant differences ( Fig. 8A ). It was the root that was most influenced, showing significant reduction in length ( Fig. 8B ) at doses as low as 5 mg/L of CuNPs (16.7 mm compared to 31.5 mm in control).

Shoot dry weight showed no significant differences, and only a tendency in reduction was observed over 25 mg/L concentration of CuNPs ( Fig. 8C ), but again a more pronounced effect was observed in root dry weight ( Fig. 8D ) which was significantly lower at 10 mg/L of CuNPs compared to CoC. A trend of gradual inhibition of root length and root dry weight was observed with increasing concentration of CuNPs as well as CoC.

Seed metabolic efficiency (SME) was calculated to reflect on the reserved food material of seeds utilized in process of germination and seedling growth in presence of CuNPs ( Fig. 9 ). No significant reduction with respect to the control was observed, but again a tendency to reduction appeared with the highest concentrations.

Most of the research on antimicrobial effects of metal nanoparticles in general have revolved around human pathogenic microbes, both bacteria and fungi, due to their importance in human health, whereas little information is available on plant pathogens.

The present investigation was carried out to know the antimicrobial potential against harmful plant pathogens and, at the same time, to address the toxicity issues of CuNPs with regard to plant and beneficial microbes. Copper is required by all living organisms as a trace element and it is part of several enzymes and electron carriers in oxidative electron transport chain. In vitro growth promotion of certain plant pathogens by CuNPs is not surprising at all and stimulation of mycelial growth by copper at high concentrations has been previously reported ( Englander & Corden, 1971 ). However, the very high growth inhibition of P. cinnamomi (oomycetes) in contrast with the other plant pathogenic fungi (Ascomycetes and mitosporic fungi) tested in the present study may be due to its pseudofungus nature and differences in cell wall and membrane composition ( Alexopoulos et al ., 2002 ). The cellulosic nature of cell wall of oomycete pathogens ( Alexopoulos et al ., 2002 ) coupled with absence of ergosterol in P. cinnamomi cell membranes ( Pérez-de-Luque et al ., 2011 ) may contribute to the higher copper sensitivity than the ergosterol-containing plant pathogenic fungi, which have chitin as a chief constituent of the cell wall ( Alexopoulos et al ., 2002 ). For example, in the yeast Saccharomyces cerevisiae, copper toxicity has been found to be dependent on plasma membrane fatty acid composition ( Avery et al ., 1996 ). In addition, physiological and metabolic differences could contribute to such susceptibility, in the same way that all those factors could be responsible for specific lack of sensitivity of oomycetes against common antifungals ( Pérez-de-Luque et al ., 2011 ). Indeed, heavy metal toxicity (which includes copper) can be due to alteration or blocking of enzymes, inhibition of macromolecule synthesis, or disruption of cellular and organellar membranes ( Arduini et al ., 1995 ; Denkhaus & Salnikow, 2002 ).

Alternaria alternata has been recently described as tolerant to high levels of copper metal ion in the medium ( Al Abboud & Alawlaqi, 2011 ), so it agrees with the results we have found in which only the highest levels of CuNPs affect its mycelial growth. The work by Al Abboud & Alawlaqial (2011) indicates how sporulation rate decreases with increasing concentration of copper in the medium, and that fatty acid composition of membranes is severely affected.

Regarding T. harzianum , there are reports about the ability of this genus to accumulate copper as a layer in the cell wall surface during the first stages of growth ( Anand et al ., 2006 ). This would explain the green colour observed in the central (and older) parts of the mycelium ( Fig. 2C ). Such a mechanism allows the fungus to remove the copper from the medium, leaving a less toxic environment for further growth ( Anand et al ., 2006 ), which probably explain why sporulation is not affected later.

An interesting phenomenon observed during the experiments was the synergistic effect between CuNPs and CoC for inhibition of mycelial growth and/or sporulation. More specific studies are needed to elucidate the exact nature of this effect, but a possible explanation could be that co-adjuvants present in the commercial formula of CoC would prevent agglomeration of CuNPs (acting as capping/stabilizing agent) and help CuNPs to penetrate into fungal cells better, increasing their toxicity. Several capping/stabilizing agents of biological origin ( Shobha et al ., 2014 ; Singh et al ., 2015 ) and chemical nature ( Guzman et al ., 2009 ) are reported in the literature.

It is known that reactivity, and consequently effectiveness of nanoparticles depends on their size. It is regarded that the antimicrobial effect of metal nanoparticles is due to their small size and high surface to volume ratio as well as the release of metal ions in solution ( Morones et al ., 2005 ). Finely divided metallic copper of size 1 µm was more effective against Phytophthora infestans , the potato blight pathogen, than that of size 2-5 µm ( Large, 1943 ). Similarly, smaller sized CuNPs are expected to be more microbicidal than their larger counterparts. Commercial grade CuNPs (APS 50 nm) were used in this experiment, which do not ensure uniform size distribution of nanoparticles. The key for increasing their effectiveness in the future may be in research with smaller and uniformly sized CuNPs.

CoC, as a copper compound, is antibacterial ( Janse, 2005 ). But in the present experiment, CoC at doses lower than 200 mg/L was found to have a growth promotion effect on P. syringae . Appearance of significantly very high numbers of P. syringae colonies was seen in CuNPs and CoC treated media plates compared to control plates in spite of inoculation with same quantity of inoculum (and therefore same number of bacterial cells in the aliquot). This reflects the fact that the control plate does not support the complete growth of all the bacterial cells inoculated, and the growth medium is at least copper deficient. The supply of copper in the form of nanoparticles and CoC promotes the growth of the bacteria.

However, CuNPs at 200 mg/L completely prevented the growth of P. syringae . It has been described that CuNPs of size 12 nm interact with the cell wall of Gram-negative bacterium E. coli and form cavities in the wall ( Raffi et al ., 2010 ). Opposite electrical charge of CuNPs make them adhere to the bacterial cell wall resulting in certain reduction reaction ( Raffi et al ., 2010 ). In P. syringae , being a gram-negative bacterium, a similar mechanism of action might be inhibiting the bacterial growth. Disruption of plasma membrane is one of the mechanisms of action of copper in microorganisms ( Englander & Corden, 1971 ; Hoshino et al ., 1999 ). It seems that binding of copper ions to the surface of bacterial cells plays an important role in the bactericidal activity ( Hoshino et al ., 2000 ). Antimicrobial effect of copper mediated by contact killing has been reported even from a dry metallic copper surface ( Grass et al ., 2011 ) or copper alloys ( Hong et al ., 2012 ).

Similarly to the results with T. harzianum , the present investigation revealed that CuNPs are not potent biocidal agents against the tested beneficial bacteria ( Rhizobium spp.), and in one case ( R. meliloti ), even promote bacterial multiplication. The differential sensitivity of the three Rhizobium spp. towards CuNPs also reflects the differences existing among them in terms of physiology. The complete inhibition of R. meliloti at 200 mg/L CuNPs while the other two species ( R. etli and R. tropici ) recorded abundant colony growth provide further scope to unravel the mechanism of action of CuNPs. In addition, commercial grade CoC also recorded different responses against the Rhizobium spp., with complete halt of R. meliloti development at 100 mg/L CoC. Therefore, application of copper compounds in soil for leguminous crops needs careful consideration, because variations in the tolerance against this heavy metal can be found in soil populations, and in the long term they can be affected, altering their presence and population structure in the soil ( Laguerre et al ., 2006 ).

Germination of wheat seeds was not severely affected by the presence of CuNPs, and the behaviour was similar to the treatment with the commercial CoC. Only the vigor index was affected by both treatments, leading to a decrease when concentration increased. This indicates that treated seeds, despite germinating at a similar rate as untreated seeds, produce less vigorous seedlings. This agrees also with the data regarding root development. It was clear that root development was more severely affected than shoot development in presence of CuNPs and CoC. This is due to the direct contact of wheat roots with the nanoparticles in the aqueous medium. Probably, the excess of copper ions released from the nanoparticles altered the root physiological elongation processes adversely leading to inhibited root growth and thereby affecting shoot development and dry weight. Indeed, inhibition of root growth by the presence of copper has been attributed to interferences with cell elongation ( Arduini et al ., 1995 ) and cell division ( Jiang et al ., 2001 ) by inducement of chromosomal aberrations, abnormal mitosis, damage to cell wall and plasma membrane integrity. It is discernible from the data presented that an inhibitory effect of CuNPs on the shoot was recorded at higher concentrations only when the roots were severely affected, so probably the effect on the shoot could be due to a low root development and not to a direct copper effect. At low concentration, shoot growth occurred at par with the control treatment.