The selection of an appropriate nitrocellulose membrane is critical for the development of a high performance lateral flow test. Each nitrocellulose membrane has unique capillary flow characteristics based on the physical attributes of the membrane and the manufacturing process that impact the flow dynamics, sensitivity, specificity, and consistency of the assay. Also important is the method used to stripe or print reagents onto the membrane.
There are a wide variety of nitrocellulose membranes commercially available from different manufacturers. Each of these membranes has specific characteristics:
- Pore Size (µm): The diameter of the largest pore in the filtration direction.
- Porosity (%): The volume of air within the membrane provided as a percent of the total volume. Porosity, also referred to as the “bed volume”, directly correlates with the amount of liquid a membrane can hold as the liquid will displace the air within the membrane.
- Thickness (µm): Impacts the bed volume, test line width, tensile strength (i.e. handling ability), and signal visibility.
- Chemical Treatment: Typically proprietary coatings that improve the binding of proteins to the membrane or to modify the hydrophobicity.
The pore size and porosity of a membrane contribute to the capillary flow rate, the rate of speed at which a sample front moves along a membrane after liquid is introduced. It is typically measured in seconds per cm, and is proportional the pore size (i.e. as you increase the pore size you will increase the flow rate). Many manufacturers will label their various membrane grades based on this capillary flow rate or the pore size (µm). In a fast nitrocellulose membrane, such as Millipore HF75, the liquid front progresses by 4 cm in 75 seconds. In a slow nitrocellulose, such as Millipore HF180, it takes 180 seconds (2.4 times longer) to cover the same distance. Slower membranes (smaller pore size/slower capillary flow rate) increase the assay run-time (i.e. the time it takes for the result to fully develop/stabilize). Run time is a critical parameter because an antibody striped at the test line will only interact with analytes in solution during the period when the sample is passing over the test line. The use of a slower membrane (e.g. HF180) will thus increase the available time for the nanoparticles, or nanoparticle-analyte complex, to bind to the test line, which in turn can increase the sensitivity. Faster membranes (larger pore size/shorter capillary flow time) reduce the incubation time between the reagents in the system, which then yields a faster but potentially less sensitive result.
Membranes of variable pore sizes and flow rates are available from a number of manufacturers including MDI, EMD Millipore, Whatman/GE, and Sartorius. Table 1 below provides examples of available membranes from these manufacturers in relation to different membrane characteristics. Keep in mind that each manufacturer treats their membranes with a proprietary mixture of surfactants and other chemicals in order to make the nitrocellulose hydrophilic. These treatments will also influence the performance of a lateral flow assay, depending on the antibody being used. Therefore, membranes with similar physical characteristics (e.g. pore size, flow rate, etc) may perform differently if purchased from different vendors. It is best to screen membranes from different vendors to account for these differences as they pertain to individual lateral flow tests.
The next step after selecting membranes for testing will be to “stripe”, or “print”, your test and control lines. The protein binding capacity of a membrane is determined by the amount of surface area available for the protein to adhere to. This surface area is a product of the membrane thickness, pore size, and porosity. In cases where you find yourself limited by the binding capacity of your current membrane, the next step would be to try a thicker membrane, a membrane with a smaller pore size, or a membrane with higher porosity. However, most proteins are compact enough (e.g. IgG antibodies), and have a small enough effective diameter, that the binding capacity of most membranes will far exceed the amount of capture reagent.
Striping of the test and control lines onto nitrocellulose membranes is typically accomplished with the use of a dispensing instrument. There are several manufacturers of reagent dispensers for lateral flow products (e.g. Kinematic, Biodot, Imagene), which may use contact or non-contact dispensing. Non-contact dispensing (i.e. spray or jetting) typically requires less volume to stripe, but can result in greater run-to-run variability. Contact dispensing systems on the other hand have relatively low run-to-run variability, but require additional volume in order to stripe the same amount of material. At nanoComposix we use an IsoFlo contact dispenser from Imagene which is shown below. You can see both the test and control lines being dispensed at a controlled volume per centimeter and total dispense distance, among other programmable parameters.
Handling of the membranes prior to and after striping is critical to the performance of the test. Prior to striping, the nitrocellulose membrane needs to be equilibrated to a controlled humidity environment of ~50% relative humidity (RH). This is because nitrocellulose membranes that are too dry may result in spotty, non-uniform lines, while nitrocellulose that is too damp will result in a widened test line that may decrease the signal intensity. It is recommended that after receiving new membranes, you expose the membrane to the “wet room” environment overnight or for a period of 8-12 hours prior to striping. Always refer to the storage and handling guidelines provided by the manufacturer. After striping, the protein still needs to be fixed onto the membrane via a drying and curing step. Curing, the exposure of a membrane to a high temperature for a short duration of time (e.g. 37°C for 1 hour), typically occurs immediately after striping in order to remove the excess water from the deposited striping solution. The membrane is then allowed to dry further by being kept in a controlled humidity environment of <30% RH to further fix the proteins to the nitrocellulose. Note that different assays may need different lengths of curing and drying for optimal performance. This should be investigated as one possible source of variability during optimization. Once the nitrocellulose has been striped and dried, it is important that the membranes be kept away from any moisture. Nitrocellulose will readily absorb any moisture in the environment after being dried, and this can result in proteins destabilizing from the striped lines. If not kept in a dry room/dry box, the test strips should be sealed in pouches with desiccant.
Three important parameters for striping nitrocellulose membranes are the reagent concentration (mg/mL), dispense speed (cm/sec), and the dispense rate (uL/cm). The values for all of these parameters will depend on the specific assay reagents, and the physical properties of the membrane being used. Typical dispense rates using a contact dispenser are between 0.5 - 1 µL/cm, which will result in a line width of approximately 1 mm, depending on the membrane. For medium and slow membranes, an initial dispense rate of 1 µL/cm is recommended. The larger pore size associated with faster membranes will allow the solution to spread further, resulting in a wider line. Decreasing the dispense rate ( e.g. to 0.8 µL/cm) is recommended to achieve the same line width. In this case, a higher concentration of the striped reagent will be required to achieve the same amount of protein per test strip. For competitive assays, an analyte-protein conjugate is dispensed at the test line rather than an antibody (e.g. drug of abuse-BSA complex). Analyte-protein conjugates tend to spread more than antibody solutions, so the dispense rate may need to be decreased even further to obtain the same line width (0.5 µL/cm). Similarly, you may also change the dispense speed of the dispensing system to achieve the same line width goals. Increasing the dispense speed is tantamount to decreasing the dispense rate, as both will lower the volume dispensed per unit of distance. Optimizing both parameters will be necessary when developing your assay. The striping concentration of the protein is also another key parameter. For sandwich assays, 1 mg/mL is a recommended starting point for test and control line antibody concentrations with typical ranges between 0.5 to 2 mg/mL. The concentration will depend on the sensitivity requirements and the affinity of the antibodies to the analyte in the sample. Typically, a very strong control line can be obtained on the low end of this concentration range (0.5 mg/mL). It is important to note that for some competitive assays, it may be necessary to stripe the test line at a concentration much lower than this (e.g. 0.1 mg/mL). The antibodies used for striping do not need to be purified from preservatives, unlike those used for conjugation, and can usually be diluted in 1X PBS buffer. 1X PBS is a standard striping buffer and is recommended for the initial optimization testing. If problems do arise from non-specific binding or lack of sensitivity when striping in PBS, you may want to examine the striping buffer which can have significant impacts on the stability of the striped protein. Titration of the salt concentration, the use of a stabilizing agent (e.g. Sucrose), or addition of a detergent (e.g. Tween-20) are all common variants used during optimization.
After striping the membranes, it is important to mark each membrane with the line location and orientation of the test and control line, which will not be visible after drying. Although this may seem trivial, it will ensure that the membrane will be placed in the right orientation when assembling your test strips. It is also important to mark any parts of the membrane where striping may have been inconsistent (such as from an air bubble in the line) so that these strips can be discarded. Otherwise, signal strength may vary significantly with no way to determine the cause.
An additional membrane blocking step may be incorporated into the assay design and can aid in improved flow, stability of the test strip, reproducibility, and blocking non-specific binding. In this process, the entire membrane may be treated with a blocking buffer, and then dried before striping. Blocking buffers can include sugars, polymers, proteins, and/or surfactants. While some developers may utilize this step, it can be time consuming during the optimization process as well as the long term manufacturing at the large scale. Alternatively, the chemicals utilized to enhance performance may be incorporated in other parts of the test strip, such as the sample pad, conjugate pad, or running buffer.
Consistent membrane striping is critical for achieving reproducible lateral flow results. If you do not have access to a reagent dispenser, NanoComposix can stripe your membranes as a custom service. Contact us at firstname.lastname@example.org for more details.