This study demonstrates the feasibility of using electrospinning to process nitrocellulose and nylon-based materials into unique membranes designed for Western (Northern and Southern) Blotting applications. We were able to generate a variety of physical states for these materials by manipulating the starting concentrations of the electrospinning solutions. For example, at low starting concentrations, nitrocellulose underwent electrospraying and deposited as 4–8 μm diameter beads, at higher starting concentrations, this polymer formed discreet sub-micron-to-micron diameter fibers. Charged nylon formed fibers over a wide range of starting concentrations, although bead formation can undoubtedly be induced by driving the initial source solution concentration below the electrospinning threshold.
For nitrocellulose, changes in average fiber diameter were most closely associated with the changes in the starting concentrations of the polymer used at the onset of electrospinning (Figure 1 and 2A). We believe this result may be explained by variables that are extrinsic to initial bulk solution properties. Fibers produced from the 120 and 140 mg/ml solutions exhibited a broad range of cross-sectional diameters (Figure 1). During electrospinning the charged jet produced from these concentrations was observed to episodically "extrude" several millimeters from the tip of syringe, dry and eject material in a non-uniform fashion into the electric field. This phenomenon can be expected to induce continual changes in solution viscosity within the electrospinning Taylor cone, producing fibers of varying sizes. In contrast, the charged jet produced from the 110 mg/ml solution was stable and less subject to drying at the tip of the syringe and produced more uniform fibers.
The changes in viscosity that occur as a function of nylon concentration in HFIP can be described over a wide range of conditions with a 1st order equation (Figure 5A). However, our analysis suggests that fiber diameter is only directly coupled to the bulk solution properties at very low nylon concentrations (Figure 5B). At high concentrations fiber diameter was not directly correlated with solution viscosity. The variables that underlie this result remain to be defined in this system; it is possible that local changes in solution viscosity at the Taylor cone contribute to this result.
Fiber size and pore size tend to track together in the electrospinning process [7,8]. This property makes it theoretically possible to tailor membranes to specific applications. For example, for slot (dot) blotting, a membrane must exhibit high surface area and be permeable to the staining and wash solutions. Electrospun materials meet these critical characteristics. A blotting membrane composed of discreet, individual nano-to-micron diameter sized fibers has an extensive surface area  available for protein binding events, a physical characteristic that can be expected to increase sensitivity and the dynamic range available to this type of assay. The interconnected nature of the pores present in an electrospun membrane can be exploited to improve the penetration of proteins and the flow through of staining and wash buffers. Membranes composed of electrospun nitrocellulose exhibited superior performance in our slot blotting experiments with respect to the parent material (Figure 3). Results with membranes composed of electrospun nylon were less consistent (data not shown). We ascribe this result to the hydrophobic nature of charged nylon; it underwent drying when the vacuum was applied to blotting apparatus. In turn, this was associated with increased non-specific binding and background noise in the staining lanes, limitations that should be amenable to correction through changes in pore size and/or the development of composite materials.
Membranes designed for conventional electroblotting must meet criteria similar to those described for a slot blotting membrane. As with slot blotting assays, membranes composed of electrospun nitrocellulose exhibited excellent dynamic range and sensitivity in this application. However, poor band resolution limits the utility of this composition (Figure 6A). We suspect that performance might be enhanced in this material through post-electrospinning processing with methods used in the paper industry designed to reduce wicking of materials along fibers. It may also be possible to manipulate performance through changes in fiber alignment, perhaps by creating alternating layers of arrayed (aligned) fibers . Electrospun nylon exhibited sensitivity comparable to the parent membrane and exhibited distinct advantages at higher protein concentrations. As noted, the white, inverse protein bands observed in the control nylon membranes can be a consequence of excess antigen and/or the use of inadequate antibody dilutions (Figure 6A). We believe this staining artifact developed from excess protein loads present on the parent membrane. Control and electrospun membranes were processed simultaneously and exposed to the same antibody dilutions, indirect evidence that we exceeded the protein capacity of the parent membrane. The extensive surface area inherent to a fibrous construct  appears to increase binding capacity and clearly functions to improve the dynamic range of the assay (Figure 6B).
Electrospun nitrocellulose is a soft, flexible material that exhibits excellent signal sensitivity, but poor band resolution. Conversely, electrospun nylon withstands manual manipulation and supports high band resolution. In attempts to combine the signal sensitivity with the band resolution properties of nylon into a single membrane we tested the efficacy of two different composite materials in our Western blotting assays. Membranes composed of electrospun nylon and electrospun fibers of nitrocellulose exhibited performance limitations similar to pure preparations of electrospun nitrocellulose. The material gave excellent signal detection with poor band resolution (Figure 6B). Once again, we attribute these results to the loft of the electrospun nitrocellulose fibers. Ultimately, this limitation may be overcome by simultaneously electrospinning from separate source solutions to produce a composite membrane composed of intermingled fibers. This type of composite should exhibited less loft than the layered membrane that we tested. As alluded to earlier in this discussion, resolution may be increased through post-processing techniques designed to limit wicking/bleeding. Membranes composed of the electrospun nylon and (electrosprayed) beads of nitrocellulose provided much better performance. This material was far more compact than the fiber-fiber composition. Signal detection was approximately 4 fold better than either the parent nylon or the electrospun variant, band resolution was excellent (compare lanes 5–9 Figure 6A and 6C).