Root exudation may also be altered after nanomaterial exposure.In addition, adsorption of nanomaterials to bacterial cell surfaces has been reported to disperse nanomaterial agglomerates.Such processes and other soil characteristics could cause temporal variations in CNM behavior within the natural soil environment, including differentially over the course of plant growth. The results of the CNM concentration-dependent agglomeration in aqueous soil extracts qualitatively explained the observed inverse dose–response trends, which deviate from typical sigmoidal dose–response relationships reported for toxicants that dissolve in soil water , but quantitative tests are not possible because of the complex soil characteristics and dynamic processes described above. In this study, with nondissolving but agglomerating CNMs, small amounts of CNMs in moist soil did not agglomerate but rather remained suspended in soil water where they were more bio-available and impactful to soil microbes and plant roots. With larger amounts of CNMs in moist soil, large agglomerates formed, which led to a sharp decrease in their bio-availability and observed impacts . Although the inverse dose–response patterns were mostly shared across CNMs, the relationships were linear for CB and fit a power function for MWCNTs . Differences in agglomeration and possibly differing toxicity mechanisms could explain the differing model fits. Our results demonstrate that not only the mass concentration and primary particle size but also the level of agglomeration may play critical roles in determining CNM effects on plants and their root symbioses in soils. In prior microbial toxicity and hydroponic phytotoxicity studies, it was recognized that nanomaterial effects would increase as nanomaterial size decreases but would decrease as nanomaterials agglomerate. For instance, antimicrobial activity was found to be higher for smaller versus larger graphene oxide sheets,while debundled, short,nft system and dispersed MWCNTs were demonstrated to have relatively higher bacterial cytotoxicity due to enhanced MWCNT–cell contact.Depicted as “nano darts”, individually dispersed single-walled carbon nanotubes were reported to induce more bacterial death than SWCNT aggregates, as dispersed SWCNTs directly damaged bacterial cell membranes.
In hydroponic studies, dispersed MWCNTs were found to have stronger effects on tomato plants than MWCNT agglomerates.Even when comparing among agglomerates, small MWCNT agglomerates exerted stronger impacts to Arabidopsis T87 cells than large agglomerates.Still, the dose–response relationship for unstudied low concentrations, which are, across the herein unstudied range of 0 to 0.1 mg kg−1, is uncertain. It is possible that the whole-plant N2 fixation potential decreased continuously with CB concentration until 0.1 mg kg−1 . Alternatively, there could be a threshold concentration somewhere between 0 and 0.1 mg kg−1, possibly close to the lowest studied dose , above which the inhibition of the whole-plant N2 fixation potential occurred but below which it did not . There is uncertainty in such untested low concentration regimes. Such uncertainty reinforces the challenges in extrapolating toxicological results from studies using only high nanomaterial concentrations to low concentration exposure scenarios, owing to influential effects of nanomaterial physicochemical structuring.We chose multi-walled carbon nanotubes and graphene nanoplatelets as two representative engineered CNMs, with industrial carbon black for comparison. CB has been commercialized for decades in the rubber and pigment manufacturing industries,with annual production of over 10 million metric tons.However, there is evidence that CB may have similar or higher toxic effects on soil bacterial communities and amphipods compared with other CNMs.Therefore, assessing whether CB affects soybean and N2 fixing symbioses and comparing how the effects differ from those of MWCNTs and GNPs are important from an environmental regulatory standpoint. MWCNTs and GNPs were purchased from Cheap Tubes Inc. ; carbon black was purchased from Dorsett & Jackson Inc. . Besides the manufacturer reported properties , CNMs were characterized by transmission electron microscopy , thermogravimetric analysis , and inductively coupled plasma optical emission spectroscopy for material morphology, thermal stability, overall purity, and metal composition, following previously reported methods.The CNMs were used as received without further purification.
Three concentrations of MWCNTs, GNPs, and CB were evaluated in this study. A sequential 10-fold dilution method accompanied by mechanical mixing was used to prepare homogenized soil and CNM mixtures as reported previously.The mixing was performed using a hand-held kitchen mixer, from the low to the high CNM concentration treatments, with the mixer cleaned between different CNMs to avoid contamination. The cleaning procedure followed guidelines recommended by the National Institute for Occupational Safety and Health for cleaning surfaces contaminated with carbon nanotubes.CNM dry powder was weighed and amended directly into soil in concentrations of 0.01, 10, and 100 g kg−1 . Each mixture was blended thoroughly using the mixer for at least 10 min. These CNM–soil stocks were then diluted ten times by the addition of unamended soil and mixing by the mixer similarly as above, resulting in concentrations of 0.001, 1, and 10 g kg−1. The dilution and mixing were repeated again to achieve the final CNM working concentrations of 0.1, 100, and 1000 mg kg−1. The CNM–soil mixtures were stored prior to planting.Bradyrhizobium japonicum USDA 110 was initially streaked from frozen stock glycerol onto solid modified arabinose gluconate medium24 with 1.8% agar in a Petri dish, then cultivated in the dark. Following incubation, several discrete colonies were dispersed into 4 mL of liquid MAG medium. An aliquot was inoculated into a 500 mL glass flask containing 100 mL of liquid MAG medium and incubated in the dark for 5 d until stationary growth phase. Aliquots of the culture were dispensed into centrifuge tubes and centrifuged , and the supernatant was discarded. Cell pellets were resuspended in a 1 M MgSO4 solution to an optical density at 600 nm of 1.0 to serve as the inoculum during seed planting. Soybean seeds were purchased from Park Seed Co. . Seeds were inoculated with B. japonicum following the method of Priester et al.Specifically, seeds were soaked in the B. japonicum inoculum for 10 min and deposited into rehydrated peat-filled seed starter pellets at 1/4-in. depth using forceps. An aliquot of the B. japonicum inoculum was dispensed into the pellet holes over the planted seed; the seed plus additional inoculum were then covered with a thin layer of the peat pellet substrate. The pellets were watered daily and incubated on a heating mat . Each planting pot was comprised of a 3 qt high density polyethylene container with bottom perforations, which was lined with polyethylene WeedBlock fabric at the bottom, and overlain by 400 g of washed gravel to allow water drainage.
A polyethylene bag punched with 40 evenly spaced 5 mm holes was placed over the gravel, and 2.3 kg of soil was weighed into each bag. Perforation of the bags allowed for water drainage, thereby preventing root rot within the soil-filled bags. Overall, there were 10 treatments, including three concentrations for each of CB, MWCNTs, and GNPs, plus a control soil without nanomaterial amendment. There were eight replicate pots per treatment. Ten days after seed sowing, 80 VC stage 59 seedlings were transplanted into potted soils. Prior to transplanting, the outside mesh of the starter pellets was removed carefully to minimally disturb the seedling roots. A central planting hole was formed in the soil, into which B. japonicum inoculum was dispensed. One seedling was inserted into the hole, and another aliquot of B. japonicum inoculum was dispensed onto the surface. Both inoculation steps were deemed necessary for adequate contact between B. japonicum and the soybean roots and thus effective inoculation. The filled transplanting hole was covered by a thin layer of soil, and the potted soil surface covered by a layer of WeedBlock fabric to minimize soil surface crusting and weed growth. A wooden support stake was inserted against the inside wall of each pot for later plant support by tying, as needed. After transplanting, the plants were grown for another 39 d to the R6 stage in the Schuyler Greenhouse at the University of California at Santa Barbara. The greenhouse climate was controlled using VersiSTEP automation under full sunlight. The indoor air temperature ranged from 15 to 34 °C,hydroponic gutter and the indoor photosynthetically active radiation fluctuated between 21 and 930 μmol m−2 s −1 from nighttime to daytime. Soil moisture sensors were inserted to a depth of 13 cm into the soil of seven pots to monitor soil volumetric water content, electrical conductivity, and temperature. Data were recorded at least twice daily using a ProCheck data display . Pots were watered to retain an average soil volumetric water content of 0.25 m3 m−3 .Midori Giant is a determinate soybean variety, which stops vegetative growth soon after flowering initiates.Also, N2 fixation will accelerate when plants initiate pod development. Therefore, plants were harvested at each of two stages: intermediate or final , aimed at capturing CNM effects on plant vegetative growth with early nodule formation, and then reproductive development with highest N2 fixation potential. Three replicate plants from each of the ten treatments were sacrificed at the intermediate harvest , and five replicates were sacrificed at the final harvest , when plants reached stage R6 .At harvest, plants were separated, above ground from below ground, by cutting the stem at the soil surface using a single edge razor blade. The above ground part was further divided into stem, leaves, and pods . Leaves and pods were counted and arranged according to their sizes, then photographed. Total leaf area and pod size were further quantified by analyzing the images using Adobe Photoshop software.Sub-samples of fresh leaves and pods were weighed and then stored for future analyses.
The remaining tissues were transferred to separate paper bags, then weighed before and after drying to determine wet and dry biomass plus gravimetric moisture content. The below ground plant parts were removed from the pot within the polyethylene bag surround. The soil in the bag was gently loosened from around the roots and nodules using a metal Scoopula , while minimizing root system disturbance. The relatively intact below ground parts, including roots and nodules, were rinsed in deionized water thoroughly to remove remaining attached soil, then air-dried. The nodules were carefully excised from the roots using a single edge razor blade and forceps as reported previously.Nodules were counted; sub-samples were weighed and refrigerated for later TEM analysis. The remaining nodules were weighed and then analyzed immediately for N2 fixation potential. Roots were dried and massed as above, to determine gravimetric moisture content and dry biomass. After N2 fixation potential measurements, nodules were also similarly dried and massed. After acquiring dry masses, all dried plant parts were archived for future analyses. Sub-samples of soil from each pot were collected and stored for future analyses. The N2 fixation potentials of root nodules were measured as nitrogenase activity by the acetylene reduction assay, according to standard methods with some modifications.Pure acetylene gas was generated by the reaction of calcium carbide and deionized water in a 1 L Erlenmeyer flask, with C2H2 collected into a 1 L Tedlar bag . Intact nodules that were freshly excised from cleaned plant roots were placed into a 60 mL syringe with a LuerLok Tip and incubated with 10% C2H2 . At 0, 15, 30, 45, and 60 min, 10 mL of the gas sample in the syringe was injected into an SRI 8610C gas chromatograph with a sample loop to measure the C2H2 reduction to ethylene over time. The GC was equipped with a flame ionization detector and a 3 ft × 1/8 in. silica gel packed column. Helium was used as the carrier gas at a pressure of 15 psi . Hydrogen gas and air were supplied for FID combustion at 25 and 250 mL min−1, respectively. The oven temperature was held constant . The C2H4 peak area and retention time were recorded using PeakSimple Chromatography Software . Chemically pure C2H4 gas was diluted by air and measured to establish a C2H4 standard curve . The C2H4 peak area values were converted to C2H4 concentrations against the standard curve and further to moles of C2H4 using the ideal gas law assuming ambient temperature and pressure. For each analysis, the moles of C2H4 produced were plotted over time, and the relationship was evaluated for linearity, then fitted by a linear regression model to calculate the C2H4 production rate. The N2 fixation potential was calculated as the C2H4 production rate normalized to the assayed dry nodule biomass.