The ability to introduce DNA into plants and achieve plant transformation has radically changed agriculture and provided a means to study fundamental processes in plants. Transformations can involve the integration of DNA into a plant genome (stable transformation) or the introduction of DNA into plant cells without the incorporation of transgenes into the genome (transient transformation) (Altpeter et al., 2016). Additionally, plastids, cellular organelles in plants with their own genomes, can be stably and transiently transformed (Maliga, 2004; Yu et al., 2020). Transient transformation is faster than stable transformation. It provides the opportunity to study the functions and regulation of genes either in single cells or within the context of plant organs (Birch, 1997; Sheen, 2001; Krenek et al., 2015). For example, reporters can be introduced to monitor transcriptional activation, the subcellular localization of proteins, and subcellular processes through the delivery of different molecular probes (Levy et al., 2018).

Three of the main methods employed for transient transformation in plants involve Agrobacteria, protoplast transfection, and particle bombardment. Agrobacterium-mediated transformation is limited to plant species, cultivars, and tissue types (roots, leaves, and seedlings) that are susceptible to this bacterium (Hwang et al., 2017). High cell densities of Agrobacteria are required to perform transient transformation (agroinfiltration), which can be harmful to plants (Karami, 2008; Li et al., 2009). For protoplast transfection, protoplasts are made via enzymatic breakdown of the cell wall after grinding up tissue. DNA is then delivered to the protoplasts via electroporation or polyethylene glycol using small vectors that are easy to manipulate. This process is advantageous because transient transformation can be achieved with a high frequency. Like Agrobacterium-mediated transformation, however, protoplast transfection is labor intensive, it requires optimization for each species (Baltes et al., 2017; Gou et al., 2020; Ren et al., 2020), and it does not allow studies of intact tissues. Finally, particle bombardment can be used to deliver DNA into a broader range of plants and tissues than agroinfiltration and protoplast transformation. The tissues to which particle bombardment can be applied include meristems, embryos, leaves, callus, stems, roots, pollen, styles, and petals (Lacroix and Citovsky, 2020). Particle bombardment, however, causes plant tissue damage when high bombardment pressures are used. Other limitations arise from the need to position plant material in the biolistic chamber or to align it with a gene gun, which limit the size of the specimen and the accessible sites (Haque et al., 2018; Mohan et al., 2019; Ozyigit and Yucebilgili Kurtoglu, 2020). Notably, all these methods affect plant physiology by inducing stress responses that may interfere with subsequent experiments using transiently transformed cells.

Nanomaterials are being developed for the delivery of biomolecules into plants because of their applicability to several different plant species and for their ability to increase transformation efficiency in planta (Kumar et al., 2020). The use of nanomaterials as vehicles for biomolecules has been widely studied in animal cells and tissues for drug delivery, but not in plants (McKnight et al., 2004; McKnight et al., 2005; Cunningham et al., 2018; Sanzari et al., 2019; Jain and Thareja, 2019; Jat et al., 2020). This may be due in part to the barriers presented by the cell wall and the fact that different plant species have varying physiologies, which need to be considered when addressing the interactions between the host environment and the nanomaterial (Pérez-de-Luque, 2017). To overcome the hurdles presented by plants, the size of the nanoparticles has been tuned, and surface modifications have been made to enable attachment of cargo to nanovehicles (Kumar et al., 2020). Some examples of these nanovehicles include: surface functionalized and pristine single walled carbon nanotubes (Liu et al., 2009; Demirer et al., 2019), mesoporous silica nanoparticles (MSNs) (Torney et al., 2007; Chang et al., 2013), gold nanoparticles and nanoparticles made of other materials coated with gold, polymeric nanoparticles (Zhang et al., 2019b), DNA nanostructures (Zhang et al., 2019), layered double hydroxide clay nanosheets (Mitter et al., 2017), magnetic nanoparticles (used for pollen magnetofection (Zhao et al., 2017)), silicon carbide whiskers (Arshad et al., 2013), and peptides as carriers for nucleic acids (Zheng et al., 2017; Jat et al., 2020). Of the nanovehicles used to deliver genes/DNA into plants, acicular shaped materials (carbon nanotubes) account for a large portion of the studies conducted. These materials have been delivered into plants through passive delivery, a needleless syringe, or with the aid of ultrasound, a magnetic field, or a gene gun (Jat et al., 2020).

A promising approach used for delivering biomolecules into cultured animal cells was the application of vertically aligned carbon nanofibers (VACNFs; McKnight et al., 2004; McKnight et al., 2005). Since these fibers are similar in scale to aphid stylets but with closed tips, they were assessed by Davern et al. (2016) for their potential to permit delivery of different sized molecules such as dyes, proteins, and dextrans to plant leaves. For this method, molecules dissolved in water were pipetted onto a leaf surface, and overlayed with the fiber array that was then gently tapped to promote penetration of the nanofibers and molecules into the leaf tissue. Importantly, the VACNFs enabled delivery of a variety of molecules into leaf tissue without causing a detectable stress response (Davern et al., 2016). Furthermore, fibers broke away from the backing and remained embedded in cells without causing loss of cell viability, providing markers for which cells potentially received molecules of interest.

In this work, we have pioneered the use of VACNF arrays for the purpose of delivering DNA-encoded probes to plants to enable transient transformation (impalefection). We show the flexibility of the method by using it to transiently transform plant organs in several species with DNA encoding fluorescent proteins that can be detected by confocal microscopy. By examining the plant with a minimally invasive method, this approach will provide a more physiologically accurate picture of signaling pathways than most other studies that rely on detached tissues.

In this work, two different VACNF chip impalefection protocols (on-plant and on-chip, Figure 1 ) were established for transient transformation of leaves, roots and fruit in a number of plant species ( Figures 2 – 6 ). VACNF chips can be used on intact plants or on organs detached from plants. This raises the possibilities of a number of interesting applications, such as studying the impact of environmental perturbation on reporter gene activity. Both the on-plant and on-chip methods rely on the fibers to make small punctures in the cell wall and membrane through an impulse force resulting from the tapping of the tweezers on the back of the chip. Using the fibers in this manner is minimally invasive and was not found to be toxic or damaging to the plant (Davern et al., 2016). Importantly, these methods do not require fibers to be surface functionalized, unlike other existing methods (Demirer et al., 2019; Jat et al., 2020; Zhang et al., 2022b), and the fibers can be used to deliver different sized vectors. With the exception of hypocotyls, all plant organs that we tested could be transformed using VACNFs: leaves (Arabidopsis, poplar, lettuce, and N. benthamiana), roots (Arabidopsis) and fruit (tomato; Figures 2 – 6 ). Importantly, VACNF-mediated transformation may be successfully used in a species, such as lettuce, for which Agrobacterium-mediated transient transformation is inefficient ( Figure 5 ).

In this study, VACNFs were produced at Oak Ridge National Laboratory CNMS through their user program, using previously described protocols (Melechko et al., 2005; Melechko et al., 2009). These publications also provide an overview and workflow of the fabrication of nanofiber arrays. VANCFs can also be produced at university clean rooms with direct current plasma enhanced chemical vapor deposition machines (Liu et al., 2010; Salmeen et al., 2015). A workflow summarizing important design, quality-control and operational steps for impalefection is provided in Figure 7 . Successful DNA delivery with the VACNFs in plants depends upon on the geometry of the fibers and the design of the chips. The fibers used herein were straight with a tapering diameter such that the tip was less than 200 nm and resembled an aphid stylet. It was found that sharp tips with smaller diameters were less damaging than blunt tips in Arabidopsis leaf epidermal cells (Forouzesh, 2012; Forouzesh et al., 2014). In their study, Forouzesh et al. (2013) used single tips (or a single fiber equivalent) for nanoindentation studies to quantify the failure stress of cell walls. If the fibers are bent, they are unable to make direct contact with the leaf tissue, hence unable to impale cells and achieve delivery. Straightness of fibers is one of the first things to check if VACNF-mediated transformation fails ( Figure 7 ). Pitch is another consideration, although in our experience it did not impact delivery success. The chips used in our experiments were designed to limit the number of fibers penetrating each cell (limit 1-2 fibers) by varying the pitch. Another factor that did prove critical for protein expression after VACNF-mediated impalefection is the timing. Most fluorescent markers were detected after 2 d. However, in lettuce, expression took longer, 4 d ( Figures 2 – 6 ). The fluorescent markers encoded by VACNF-delivered DNA were detected in expected localizations: endomembrane compartments and vacuoles for AALP:GFP (Sohn et al., 2003; Kim et al., 2005; Kang et al., 2014), cytosol and nuclei for free mCherry and YFP (e.g. Kang et al. (2014)).

Both large (binary vectors utilized for Agrobacterium-mediated transformation) and small vectors (previously used for protoplast transformations) can be successfully delivered and expressed in plants ( Figures 2 – 6 ). Smaller plasmids are easier to manipulate and tend to have a higher transient transformation efficiency in protoplast transfections and particle bombardment than the larger vectors used for Agrobacterium-mediated transformations (Sainsbury et al., 2009). With fiber delivery, results using larger and smaller vectors were similar. Therefore, constructs designed for use with Agrobacterium can be successfully used with VACNFs.

Nanomaterials are increasingly being used for transient transformation. The most recent methods of nanoparticle-mediated transformation or delivery of nucleic acids to plants have relied upon using pressure infiltration to deliver nanoparticles into plants via a needleless syringe (Zhang et al., 2019; Demirer et al., 2019; Izuegbunam et al., 2021; Zhang et al., 2022). Together, these nanoparticle-mediated methods have enabled delivery of nucleic acids and other molecules into Arabidopsis, tobacco, onion, maize, wheat, rapeseed, mustard, carrot, cotton, tomato, spinach, arugula, watercress, cowpea, chamomile, barley, and moss (Kumar et al., 2020; Jat et al., 2020; Izuegbunam et al., 2021). Roots can take up nanoparticles through passive diffusion and plant leaves can easily be transformed with passive infiltration of nanoparticles entering through the stomata (Hubbard et al., 2020). To the knowledge of the authors, only one study previously used nanoparticles to transform other parts of a plants that were not leaves or roots. Specifically, in Izuegbunam et al. (2021), the authors used needleless syringes to deliver nanorods with anchored DNA to developing seed tissues in Arabidopsis, field mustard, barley, and wheat. This method, like the aforementioned nanomaterial-mediated delivery of DNA (Demirer et al., 2019; Jat et al., 2020; Zhang et al., 2022b), requires the conjugation of DNA to nanorods and subsequent release of DNA once inside cells.

In summary, we have demonstrated the utility of using vertically aligned carbon nanofibers to overcome the challenges of crossing plant cell walls and plasma membranes to deliver DNA into cells. There is no release of cargo necessary, since nanofibers facilitate delivery through mechanical means into the plant, probably by forming transient channels around the fibers through which material can enter cells (Davern et al., 2016). The level of transgene expression readily allowed detection of several fluorescent reporters. Previously, VACNFs were successfully used to deliver proteins to plants cells (Davern et al., 2016). In that study dextran and proteins were detected in plant cells as early as 5 minutes after introduction via VACNFs. Labeled peptides entered vascular tissues which allowed their distribution throughout the plant (Davern et al., 2016). In the present work, we have demonstrated the value of VACNF chips in delivering DNA into leaves, roots, and fruit, with the minimally invasive impalefection allowing subsequent expression of the intended gene products. Looking forward, VACNF chips could be utilized to deliver multiple types of biomolecules together (e.g., gene editing reagents) and for stable transformation via delivery to a meristem, plant cell culture, or plants undergoing early organogenesis.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

JM, JJ, JG, RS, and JM-F conceived the research and designed the experiments. JM, SR, and DH designed and fabricated the carbon nanofibers. JM and JJ performed the experiments and analyzed the data. JM and JJ wrote the manuscript with input from JG, RS, SR, DH, and JM-F. All authors contributed to the article and approved the submitted version.

This work has been funded by the Bioimaging Science Program, U.S. Department of Energy, Office of Science, Biological and Environmental Research, DE-SC001914 and the United States Department of Agriculture, 2021-67013-34835. JMM was supported by the United States Department of Agriculture: National Institute of Food and Agriculture: Agriculture and Food Research Initiative Predoctoral Fellowship 2021-67034-35167.

We thank Dr. Inhwan Hwang (Pohang University of Science and Technology) for CaMV p35S: AALP:GFP DNA and Dr. Jiyoung Lee (Korea Research Institute of Bioscience and Biotechnology) for p35S-mCherry constructs. We thank John Zdenek for assistance with plant growth and maintenance. We thank Mindy Clark from Oak Ridge National Laboratory and Dr. Gail Taylor from the University of California, Davis for Populus trichocarpa (Nisq-1 genotype). Nanofiber arrays were fabricated at the CNMS, which is a Department of Energy Office of Science User Facility (Proposal ID : CNMS2019-103). Support from CNMS is awarded through a peer-reviewed proposal system and is provided at no cost to successful applicants who intend to publish their results (http://www.cnms.ornl.gov/user/becoming_a_user.shtml). We thank the CNMS for assistance with the production of nanofiber arrays. We thank Dr. John Caughmen, Dr. Timothy McKnight, and Travis Bee for critical discussion about experimental design.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.