IVD Technology Magazine
IVDT Article Index

Originally published September, 1998

The role of membranes in molecular diagnostics

With molecular diagnostics on the way in, some say that membranes are on the way out. But nothing could be further from the truth.

In the evolution of IVDs, membranes and microporous materials have long occupied a position of central importance. The use of membranes made from such materials as nitrocellulose and nylon made possible the development of immunoassays based on the interaction of antibodies and antigens, and enabled those tests to achieve the market dominance they enjoy today. Membranes have also made possible the refinement of immunodiagnostics into lateral-flow, point-of-care assays—familiar today in the form of home-use pregnancy test kits.

But with the commercialization of molecular diagnostics gradually unfolding, some analysts have suggested that the role of membranes is destined to diminish. Immunoassays based on antibody-antigen interactions, they say, will be replaced by newer molecular diagnostics that target nucleic acids. In the process, membranes will be displaced by such futuristic technologies as microfabricated DNA chips.

Figure 1. Applications for nylon 6,6 membranes in molecular biology and diagnostics.

Looking at current trends, however, it is difficult to see that such predictions are likely to come true. In fact, it seems even more likely that the use of membranes and microporous materials will increase as molecular diagnostics move into the marketplace. Membranes already have an interesting history of accomplishment in the realm of molecular diagnostics, including significant crossovers from molecular biology research laboratories to commercial diagnostics (see Figure 1). When the first Western transfers from polyacrylamide gels were developed, for instance, nitrocellulose membranes were selected as the medium of choice for binding the proteins transferred by the process.^1,2 Nitrocellulose membranes were also the original membranes selected for use when Southern transfers were developed. In that application, nitrocellulose has now been largely replaced by supported nylon 6,6 membranes because of the latter's easy DNA immobilization, sensitive immunodetection properties, and durability during repeated probe stripping.^3,4

Today, researchers are using membranes and microporous materials in molecular applications such as DNA arrays for diagnostics and high-throughput screening (HTS) formats for drug discovery. In the future, it is likely that membranes will find additional uses in fields well beyond those for which they were originally intended.

This article examines some of the key molecular applications for which membranes are particularly suited, with notes on selecting the appropriate membranes for each process (see Table I).

Sample Preparation

Broadly defined, sample preparation is the selective partitioning of a complex sample, or the preparation of a specific analyte or component of the sample for subsequent testing. Common examples include the separation of plasma from whole blood, removal of heparin from plasma, DNA isolation, and white blood cell preparation. Many of the new diagnostic technologies that are being developed require analytes to be selectively prepared in some manner.

Researchers often use centrifugation to accomplish many of the steps required to prepare a sample for testing. But while centrifugation may be suitable for benchtop research, it is not a process that is easily automated. Microporous membranes, on the other hand, lend themselves to automated handling and detection schemes. Several types of such materials can be used to separate plasma from whole blood or to prepare DNA or white blood cells for further testing.

A number of microporous materials are available for separating plasma from whole blood.^5,6 Originally designed to rapidly separate plasma from small quantities of whole blood for point-of-care immunodiagnostics, these materials can be thought of as a noncentrifugal method of blood separation.

In practice, plasma easily wicks from some blood separation materials (but not all) onto a wide variety of solid supports, making the separation media useful in preparing samples to be tested for bacteria, cholesterol, and a variety of plasma analytes. Such materials may also be useful in automated clinical screening for extracellular viruses such as the hepatitis viruses and HIV.

Preliminary experiments using the Hemasep V blood separation medium suggest that this material can be used as a noncentrifugal method of extracellular virus preparation.⁷ The concept is simple. A blood separation material in contact with a low-protein-binding medium allows plasma to be quickly separated from whole blood (see Figure 2). The plasma then transfers to a secondary low-protein-binding solid support, and is incorporated into a polymerase chain reaction (PCR) instrument designed to handle a planar surface.

Figure 2. Using Hemasep membranes to prepare a virus sample from whole blood.

Experimentally, whole blood spiked with M13 virus was applied to Hemasep V medium, and plasma was transferred to a variety of low-protein-binding, secondary solid supports that were then subjected to PCR amplification. The results demonstrated that this blood separation material can be used to prepare extracellular viral DNA for amplification without the use of centrifugation (see Figure 3).

Figure 3. PCR directly from diagnostic media placed in contact with Hemasep V medium. A 1:25 dilution of lysate was added to 40 ml of fresh human blood and spotted onto Hemasep V medium. A 3-mm piece of medium from the plasma zone was added directly to a PCR reaction (d), or a diagnostic membrane was placed in contact with Hemasep V to collect plasma. Secondary media included Premium Release membrane (e), LoProsorb medium (f), and LoProdyne LP nylon 6,6 membrane (g). Membrane samples (3 mm) containing the separated plasma were also placed directly into PCR reactions. The PCR controls were (a) M13 DNA, 100 ng; (b) lysate, 10⁹ particles; and (c) no DNA or bacteriophage.

The idea of performing PCR directly from separation membranes is not unusual; the literature abounds with examples of studies—successful and unsuccessful—performed to evaluate this approach.^8,9 Success appears to depend in large measure on the membrane used and whether it inhibits the amplification reaction.

Other membranes are useful for separating leukocytes (white blood cells) from whole blood samples for further analysis. These materials function in the same way as leukocyte reduction filters that are widely used in transfusion medicine, selectively binding white blood cells while allowing the majority of red blood cells to pass unimpeded. Leukosorb medium, for instance, is a flat-sheet, fibrous material that can remove approximately 90% of white blood cells from a whole blood sample. Vertically stacking layers of the material increases their removal efficiency.

Solid supports that selectively bind leukocytes can also be used to prepare genomic DNA from whole blood samples, or to permit examination of surface receptors on trapped leukocytes. In addition, a number of viruses, such as cytomegalovirus, are localized primarily in white blood cells.¹⁰ Leukocyte-binding solid supports can be used to trap white blood cells so that such intracellular viruses can be submitted for further analysis.

Membrane-Based DNA Assays

There are a number of reasons for employing membranes in DNA-based, molecular assays. In addition to their familiarity and established performance record, membranes often offer qualities that cannot be matched by other substrates, including the following:

• Quicker assay results.
• Simplified screening methods.
• Compatibility with a variety of established DNA and protein detection systems.

Molecular assays that incorporate membranes as solid supports currently exist in several different forms. For example, Orgenics, Inc. (Yavne, Israel), markets a paper chromatography hybridization assay (PACHA) that uses nitrocellulose membranes.¹¹ This rapid assay employs a number of target-specific oligonucleotide probes, which are immobilized as stripes on the membrane. To perform the assay, DNA is amplified by PCR and biotinylated nucleotides are incorporated during the reaction. This biotinylated DNA is then applied to the membrane, where it is drawn by capillary action over the immobilized probe stripes. Sequence-specific hybridization occurs at each probe stripe where the biotinylated DNA encounters its molecular counterpart, while nonhybridized probes migrate past the probe stripes. Finally, streptavidin-alkaline phosphatase conjugates are added to develop signal from the hybridized probes.

In another application, a company has studied the possibility of detecting foodborne pathogens using DNA probes in a dipstick format.¹² Although the study used plastic strips as solid supports, membranes would also have worked and could become the support of choice for such tests. The test employed a mixture of probes containing a DNA sequence designed to recognize a specific pathogen, and a separate polydeoxyadenylate tail designed to recognize the polydeoxythymidylate-coated solid support.

DNA Arrays

High-density microarrays are becoming more widely used for DNA sequencing, genetic analysis, and drug discovery.¹³ Although manufacturers are exploring a wide range of substrates for such arrays—including microfabricated DNA chips—membranes are also being used. Nylon membranes are used for DNA sequencing by hybridization, and are also available as solid-phase supports that incorporate arrays of bacterial colonies or tissue-specific mRNAs.¹⁴ Membranes that were originally designed for use as lateral-flow materials in immunoassays are also finding utility as the solid supports for new array technologies.¹⁵

An increasing number of companies have begun to offer commercial DNA arrays as well as "libraries" of bacterial clones stored on membranes. Most of these new products are based on standard hybridization principles. Such commercial membrane-based arrays can be used to analyze gene expression, investigate the mechanisms of diseases, or identify useful compounds for drug discovery. Clontech Laboratories, Inc. (Palo Alto, CA), for instance, offers membrane-based nucleic acid arrays that are designed to help monitor gene expression. The company's arrays use positively charged nylon membranes as their supports. Other companies in this field include Research Genetics (Huntsville, AL), which offers a high-density array of more than 5000 genes on a membrane 5 X 7 cm, and Display System Biotech (Vista, CA), which offers membrane-based arrays for probing gene expression.

The Correct Membrane?

Although many companies are using membranes for DNA sequencing or drug discovery applications, few have expanded their design parameters beyond the use of nitrocellulose and radioactive methods of signal detection. To optimize the performance of their products, however, manufacturers need to ensure that they are using the appropriate membranes and detection methods for the task. Existing nylon 6,6 membranes, for instance, can be used to detect DNA via colorimetric, fluorimetric, and chemiluminescent methods (see Figure 4).¹⁶

Figure 4. The flexibility of Biodyne Plus membranes in molecular applications is shown in these 384-spot DNA arrays, which were produced using three different chemical substrates. For this test, 200 pg of Lambda Hind III DNA was applied to 384 spots on Biodyne Plus membranes. The membranes were hybridized with 50 ng/ml DIG-labeled probe according to instructions. After hybridization, the membranes were washed and blocked, and incubated with anti-DIG antibody conjugated to alkaline phosphatase. Signal was developed using three different alkaline phosphatase substrates: (a) BCIP/NBT, for which colored signal forms directly on the membrane; (b) chemiluminescent dioxetane-based substrate, with signal developed on film; and (c) precipitating chemifluorescent substrate, shown as a fluorescent scan from a Molecular Dynamics Fluorimager.

When designing a nucleic acid–based, molecular assay that will employ a solid support, a nylon 6,6 membrane is a good choice. However, nylon membranes are not all alike. Nylon formulations and manufacturing processes differ, resulting in membranes that behave differently in identical assay conditions. Although both Biodyne B and Biodyne Plus membranes are positively charged, for instance, they are manufactured by different methods. The result is that under identical assay conditions one membrane provides excellent sensitivity and low background, while the other is unreadable (see Figure 5). Other nylon 6,6 membranes also offer different surface chemistries, resulting in differing signal-to-noise ratios. Because of these differences, it is difficult to extrapolate the data and protocols from one nylon membrane to another.

Figure 5. The differing surface chemistries of nylon 6,6 membranes can result in varied signal detection quality and background levels. Note the difference between membranes (b) and (c), two positively charged materials (Biodyne Plus and Biodyne B, respectively) made by different methods. Also shown are Biodyne A (a), Biodyne C (d), and LoProdyne LP (e).

Superior performance can be achieved when product designers optimize both their choice of membranes and their DNA detection protocols. Following are some areas that developers should consider when selecting membranes for molecular applications.

Membrane Characterization. Manufacturers can greatly reduce assay development time by using existing membranes. Because they are often well characterized with current protocols and detection systems, existing membranes require less research and development on the way to a finished product. Although there will be development costs associated with the assay itself, existing membranes require no additional development and hence no additional cost.

Nylon 6,6 is a good example of how membrane characterization can work to a developer's advantage. Nylon 6,6 membranes are well characterized with many DNA-based protocols and commercial detection systems, and many protocols exist for optimizing DNA and protein detection on nylon 6,6 membranes.¹⁷ Much of the membrane characterization has been conducted by membrane manufacturers, but even more work has been reported by independent researchers in peer-reviewed articles. In addition to reports about the use of nylon 6,6 membranes with specific detection systems, there is also a fair amount of information about the use of such membranes for DNA sequencing, DNA detection in multiplexed sequencing, and reverse dot-blots for clinical diagnostics, to name just a few applications.^18–22

Surface Chemistries. A major advantage of current nylon membranes is the variety of surface chemistries available: neutral membranes, positively charged membranes, and negatively charged membranes are each suited for some application in molecular diagnostics. To determine what type of surface chemistry is best suited for a new molecular assay, developers should take into account both the requirements of the specific protocol and the method of DNA detection that will be used. Usually, it is the complex interaction of reagents used in the detection process that dictates a developer's choice of membrane surface charge (see Table II).

Although the term surface chemistry is used, the chemistries permeate the entire three-dimensional lattice of the membrane. In addition, the charge is integral to the membrane and does not wash away. It is often stated that nitrocellulose membranes bind molecules by means of hydrophobic interactions, while nylon membranes bind molecules by means of ionic or electrostatic interactions.²³ However, the binding mechanism of nylon membranes is actually more complex than this. Although the membrane's charge component plays a role, binding to nylon membranes occurs primarily by means of hydrophobic interactions (see Figure 6). This mechanism is suggested by the fact that, under similar conditions, positively, negatively, and neutrally charged nylon 6,6 membranes bind approximately the same amount of DNA or protein.

Figure 6. Protein binding to nylon 6,6 membrane, showing the association between the protein and the membrane.

Durability for Automated Processes. Some nylon 6,6 membranes are manufactured in both supported and unsupported versions. Supported versions are more durable, and therefore more suited to applications for which the membranes must undergo additional processing or assembly, or where the assay itself is to undergo automated processing. Unlike laminated materials, the support in such membranes is internal, with nylon deposited on both surfaces (see Figure 7). This results in membranes that are durable, with support that is invisible to the assay, and without a particular orientation (either surface can be employed).

Figure 7. Scanning electron micrograph showing the microstructure of Biodyne nylon 6,6 membranes.

Detection Sensitivity. With the development of nonradioactive detection technologies, the use of radioactive methods is gradually becoming less common in the field of molecular diagnostics. The three most prominent nonradioactive detection technologies now in use are colorimetric, chemiluminescent, and fluorescent methods.

The displacement of radioactive methods depends very heavily on the ability of the newer technologies to offer similar levels of sensitivity. Today, membrane-based DNA assays using chemiluminescent or fluorescent methods with signal amplification reagents can achieve detection levels comparable to radioactive detection. Chemiluminescence, in particular, is gaining ground in clinical diagnostics for screening applications ranging from antibodies to vitamins.²⁴

The key to optimizing signal detection is to match the type of membrane in use to the type of detection method that will provide the best possible results. When used in conjunction with the appropriate detection method, nylon membranes can provide excellent signal-to-noise ratios (high signal and low background). In the extremely demanding field of forensic science, for instance, federal laboratories use a neutrally charged nylon membrane for nonradioactive, chemiluminescent detection, but have found that a positively charged nylon membrane works well with radioactive detection methodologies.^25–27

New Membrane Development

Existing membrane technologies provide a range of alternatives for the developers of molecular diagnostics and HTS applications. As these developing fields mature, however, new membranes will need to be developed to meet their specific requirements.

For molecular diagnostics, membranes will be required to show improved detection capabilities when used with nonradioactive detection methods. Also, new affinity separation membranes need to be developed. These media will be used during sample preparation to selectively enhance or remove target molecules.

In HTS laboratories, the number of compounds that are screened continues to increase, creating a corresponding need for arrays of increasing density. Future membrane developments to meet this need should include optimized materials with superior cosmetic characteristics to permit higher dot densities. Such materials will also be required to have improved signal-to-noise ratios when used with chemiluminescent and fluorescent detection systems.

Conclusion

As new DNA detection technologies vie for acceptance in the marketplace, developers should take note not to overlook the potential of membranes and microporous materials. Such materials have an established record of successful applications in the laboratory for DNA and protein detection and analysis. The combination of a well documented track record, consistent membrane performance for DNA detection, and low cost all combine to help ensure that microporous membranes will increasingly be used as enabling technologies for molecular diagnostics and HTS applications.

References

1. Towbin H, Staehelin T, and Gordon J, "Electrophoretic Transfer of Proteins from Polyacrylamide Gels to Nitrocellulose Sheets: Procedure and Some Applications," Proc Natl Acad Sci, 76:4350–4354, 1979.

2. Burnette W, "Western Blotting: Electrophoretic Transfer of Proteins from Sodium Dodecyl Sulfate-Polyacrylamide Gels to Unmodified Nitrocellulose and Radiographic Detection with Antibody and Radioiodinated Protein A," Anal Biochem, 112:195–203, 1981.

3. Dubitsky A, and Defiglia J, "Stripping of Digoxigenin-Labeled Probes from Nylon Membranes," BioTechniques, 19(2):210–212, 1995.

4. Noppinger K, Duncan G, Ferraro D, et al., "Evaluation of DNA Probe Removal from Nylon Membrane," BioTechniques, 13(4):572–575, 1992.

5. Alter J, "One-Step Separation of Plasma from Whole Blood for In-Vitro Diagnostics," Gen Eng News, 16(5):28, 1996.

6. Alter J, "Single-Step Vertical Plasma Separation of Whole Blood for Tests and Sample Prep," Gen Eng News, 16(20):30 1996.

7. Alter J, and Seeley K, "Automation of Viral Detection Utilizing Solid Support Plasma Separation from Whole Blood," LabAutomation '98 (abstract), 146, 1998.

8. Oyofo B, and Rollins D, "Efficacy of Filter Types for Detecting Campylobacter jejuni and Campylobacter coli in Environmental Water Samples by Polymerase Chain Reaction," App Environ Microbiol, 59(12):4090–4095, 1993.

9. Yourno J, "Direct Polymerase Chain Reaction for Detection of Human Immunodeficiency Virus in Blood Spot Residues on Filter Paper After Elution of Antibodies: An Adjunct to Serological Surveys for Estimating Vertical Transmission Rates Among Human Immunodeficiency Virus Antibody-Positive Newborns," J Clin Microbiol, 31(5):1364–1367, 1993.

10. Rice GPA, Schrier RD, Oldstone MBA, "Cytomegalovirus Infects Human Lymphocytes and Monocytes: Virus Expression Is Restricted to Immediate-Early Gene Products," Proc Natl Acad Sci USA, 81:6134–6138, 1984.

11. Reinhartz A, Alajem S, Samson A, et al., "A Novel Rapid Hybridization Technique: Paper Chromatography Hybridization Assay (PACHA)," Gene, 136:221–226, 1993.

12. Groody E, "Detection of Foodborne Pathogens Using DNA Probes and a Dipstick Format," Mol Biotech, 6:323–327, 1996.

13. Regalado A, "The DNA-Chip in Diagnostics," Start-Up, (September):18–25, 1996.

14. Drmanac S, and Drmanac R, "Processing of cDNA and Genomic Kilobase-Size Clones for Massive Screening, Mapping and Sequencing by Hybridization," BioTechniques, 17:328–336, 1994.

15. Stimpson D, BioTechniques, November 1998, in press.

16. Dubitsky A, "DNA Arrays on Nylon Membranes," LabAutomation (abstract), June 1998.

17. Dubitsky A, "Blocking Strategies for Nylon Membranes Used in Enzyme-Linked Immunosorbent Assays," IVD Tech, 3(4):53–59, 1997.

18. Weiss N, Eggersdorfer I, Keller C, "Multiplex-PCR-Based Single-Strand Conformation Polymorphism Protocol for Simultaneous Analysis of Up to Five Fragments of the Low-Density-Lipoprotein Receptor Gene," BioTechniques, 20:421–429, 1996.

19. Cherry JL, Young H, Di Sera LJ, et al., "Enzyme-Linked Fluorescent Detection for Automated Multiplex DNA Sequencing," Genomics, 20:68–74, 1994.

20. Martin C, Bresnick L, Juo R-R, et al., "Improved Chemiluminescent DNA Sequencing," BioTechniques, 11(1):110–112, 1991.

21. Zhang Y, Coyne MY, Will SG, et al., "Single-Base Mutational Analysis of Cancer and Genetic Diseases Using Membrane Bound Modified Oligonucleotides," Nuc Acids Res, 19:3929–3933, 1991.

22. Schollen E, Vandenberk P, Cassiman J-J, et al., "Development of Reverse Dot-Blot System for Screening of Mitochondrial DNA Mutations Associated with Leber Hereditary Optic Atrophy," Clin Chem, 43(1):18–23, 1997.

23. Harvey M, Audette C, and McDonogh R, "The Use of Microporous Polymer Membranes in Immunoassays," IVD Tech, 2(3):34–40, 1996.

24. Kricka L, "Chemiluminescence Takes Clinical Diagnostics to a New Level," Advance, (May): 10–13, 1997.

25. Giusti A, and Budowle B, "A Chemiluminescence-Based Detection System for Human DNA Quantitation and Restriction Fragment Length Polymorphism (RFLP) Analysis" Appl and Theoretical Electrophoresis, 5:89–98, 1995.

26. Giusti A, and Budowle B, "Effect of Storage Conditions on Restriction Fragment Length Polymorphism (RFLP) Analysis of Deoxyribonucleic Acid (DNA) Bound to Positively Charged Nylon Membranes," J Forensic Sci, 37(2):597–603, 1992.

27. Benzinger EA, Shirley RE, Riech AK, et al., "Time Course and Inhibitors of Hae III Digestion in the Forensic Laboratory," Appl and Theoretical Electrophoresis, 4:179–188, 1995.

Jason M. Alter, PhD, is vice president for marketing at Pall Specialty Materials, a division of Pall Corp. (Port Washington, NY). The author wishes to thank Andrew Dubitsky for the figure depicting DNA arrays on nylon 6,6 membranes (Figure 4).

Photo Courtesy Pall Corp.