Development of Chip-compatible Sample Preparation for Diagnosis of Infectious Diseases

Marion Ritzi-Lehnert

Expert Rev Mol Diagn. 2012;12(2):189-206.

Point-of-care Testing Using lab-on-a-chip Systems

Cardiovascular diseases and cancer are currently the highest causes of death in the world, followed by infectious diseases. Infectious diseases are manifold – for example, sepsis, respiratory diseases such as various strains of influenza or tuberculosis, other bacterial infections such as enterohemorrhagic Escherichia coli, viral infections such as HIV and hepatitis B virus, parasitic infections like malaria, and so on. The large number of patients and the plethora of different infectious diseases demonstrate the need for new fast and efficient point-of-care analysis tools, especially in cases where the course of the disease is fast and severe. Therefore, the development of cheap and fast reliable technologies for diagnosis and therapy monitoring is one of the most urgent topics in modern medical science. Bringing the diagnostic device to the point of care and comprising all steps from specimen preparation to signal readout will help to accelerate determination of patient health status and specific antibiotic prescription.

Point-of-care tests have to be fast and easy to handle, which inherently conflicts with the inevitable need for high sensitivity and specificity to reliably detect and identify relevant pathogens. Simultaneous testing for a variety of pathogens and their antibiotic resistance from the original sample is advantageous. Instruments fulfilling these needs are expected to be a key enabler of improved clinical outcomes and more successful standards of care. Patient mortality, development of microbial antibiotic resistance and ultimately healthcare costs will be reduced.^[9–12] Improved health economics due to optimization of antibiotic prescription, better infection control practices, as well as reduction of resistance spreading and of clinical visits or hospital stays can be achieved.

Currently, a wide range of specialized, easy-to-handle kits are available on the market.^[13] The most prominent examples are the lateral-flow dipstick tests, such as those used for the discrimination of bacterial or viral infection. Although the protocol steps are very simple, the entire workflow (from sample to result) often demands time-consuming manual handling. In fact, most analyses need more complex process steps than can be realized within a capillary-driven fleece. As a result, these manual assays have been transferred to more convenient automated solutions (e.g., pipetting robots). This effort also led to the development of lab-on-a-chip (LoC) systems, which can be seen as miniaturized automated systems. These systems usually consist of a microfluidic-based disposable chip (most often made of polymer material) and an operating device. Applying microtechnology, microfluidics, has the potential to make a major contribution to decentralizing and simplifying medical/diagnostic testing. LoC systems that integrate several laboratory functions within a single polymer substrate give rise to new perspectives with respect to early diagnosis, disease therapy and monitoring of various diseases.^[9,11,12,14] It is predicted that by 2015, many in vitro diagnostic analyses, both laboratory-based and those designed for point-of-care testing, will use some form of miniaturization and '(bio) chip' technique.^[15,16] These systems will include both the sample preparation, during which target molecules are extracted directly from the original raw sample, and the NA-based and/or immunological detection and identification steps, in order to build an integrated system implementing the whole laboratory procedure 'from sample to result' in a fully automated manner.

Besides small sample volumes, commonly mentioned advantages of LoC systems are reduced analysis time, reagents consumption and contamination risk. Additional benefits are enhanced reproducibility and reliability, as well as the ease of use by unskilled operators outside the well-defined laboratory setting.^[7,14,16,17] In addition, a major driver in the development of LoC systems is the reduction in costs per analysis and per single test result they afford. However, critical factors for the marketability of such devices also include a high significance and reliability of the results, as well as sensitivities and specificities that satisfy stringent clinical requirements. At this stage, many microfluidics-based concepts and technologies fail to meet these demands.^[18–28] Apart from simple dipsticks or lateral flow-based devices and a few multiplex reverse-transcription (RT)-PCR applications, few LoC devices have found their way into clinical practice, despite the huge effort during the last two decades.

1 2 3 4 5 6 7 8 9 10 11 12 13

Next Section

Abstract and Introduction
Conventional Diagnosis of Infectious Diseases
Point-of-care Testing Using Lab-on-a-chip Systems
Sample Preparation: The Challenging Task
On-chip Sample Preparation
Sample Transfer (World-to-chip Connection)
Plasma Generation
Concentration of Pathogens, Elimination of Contaminants & Volume Reduction
Lysis
NA Isolation
Integrated Systems
Commercially Available Systems
Expert Commentary & Five-year View
References
Sidebar

Table 1. Different chip-based plasma generation techniques.
Technology	Volume (µl)	Time required (min)	Plasma generated	Advantages	Disadvantages	Ref.
Series of constrictions and bifurcations	50	Not given	100% with hematocrit level ≤30%	Flexible volume, parallelization possible	Dilution needed, risk of blockage	[91]
Suction through membrane	10–50	<1	2–22 µl	Flexible volume, parallelization possible	Risk of hemolysis and blockage	[96]
Egg beater (centrifugation)	~100	10	~50 µl	Flexible volume, parallelization: ≤20 samples/run	Complex integration	[98]
Rapid flow separation	0.5–1.5	<1	100% with hematocrit level of 15–60%	Additional determination of hematocrit	Large space required for separation of very low volume	[101]
Centrifugal chip module	≤300	5–10	~40–80%	Flexible volume, parallelization possible	Complex integration	(Figure 1B)

Table 2. Various chip-based concentration methods.
Technology	Sample	Reduction/concentration	Time required	Result	Advantages	Disadvantages	Ref.
Antigen-specific bead array or size-specific membrane	Bacteria, spores, T lymphocytes in 30 µl finger-tip blood or saliva	Not given	5 min (Bacillus globigii) to <20 min (T lymphocytes) (complete analysis)	LOD: 50–600 cells/µl (T lymphocytes), detection limit ~500 B. globigii, 2× sensitivity of ELISA and LOD values one order of magnitude lower	Programmable, flexible assay design	Only small sample volumes processable	[28]
Soft inertial force induced migration (lift forces and secondary flows) 18 µl/min	Escherichia coli in blood samples (5–10% v/v)	300-fold enrichment	Not given	99.87% recovery of bacteria, 62% separation recovery	Similar cell sizes distinguishable	Need for proper controlled fluid conditions, high dilution of sample, slow, prioritization on purity, recovery or throughput needed	[37]
MACS® on fixed bead bed in a bifurcated channel system	10 cfu E. coli in 20–50 µl PBS, 100–1000-fold background	Down to 10 µl	Not given	70% capture efficiency, 0.2 cfu/µl detection limit (0.3% Staphylococcus aureus capture)	Combination of volume reduction and target isolation, use of permanent magnets allows for easy bead handling	Limited volume range (10 µl too low, 100 µl too high), precoating needed to avoid bubbles and sticking of beads/cells	[36,58]
Evaporation-induced meniscus-dragging effect	100 µl, 100 cfu MSSA/µl in water	1:2000	15 min	Recovery concentration >85% for <10⁴ cfu/µl	High reduction factor, efficient concentration – theoretically no loss	Precleaning step needed, high power consumption	[83]
MACS (microbeads)	MCF-7 cells spiked in 5–7.5 ml blood	Down to 10 µl	1 h	Detection limit: two spiked cells	High recovery rate, highly reproducible	High number of magnetic beads needed, no time reduction	[102]
Antibody-antigen release from immobilized antibodies by UV irradiation of photo-cleavable crosslinkers	10 µl whole blood spiked with 2 ng/ml PSA and 15 U/ml CA15.3	1:2	<20 min	0.0318–53 ng streptavidin/ml, initial signal increase at low relative analyte concentrations is directly proportional to species concentration	Label-free multiplex detection	Only small sample volumes processable, optical accessibility for complete binding area	[103]
Inertial microfluidics 20 ml/h + magnetic beads	10 ml blood + 10⁶ Candida albicans/ml	Not given	5 min: 48 ± 4% bound 30–60 min: 85–90% bound	80% clearance of fungi, pathogen binding efficiency ~96 ± 1%	High sample throughput, design allows channels to be densely arrayed	Blood loss and blood dilution, accumulation of magnetic beads at channel walls	[104]
Micropillar surface with organosilane	500 µl 5 × 10⁶ cfu E. coli/ml in 1:1 (v/v) urine/sodium acetate buffer	~1:10	~10 min	48 ± 10% capture efficiency, LOD: 102 cfu E. coli/ml in acetate buffer spiked into urine samples 1:1 (v/v) ratio	High surface-to-volume ratio (0.152 µm⁻¹) by hot embossing	Complex polymer imprinting, organosilane coating and bonding	[107]

CA15.3: Cancer antigen 15-3; cfu: Colony-forming unit; LOD: Limit of detection; MSSA: Methicillin-sensitive Staphylococcus aureus; PBS: Phosphate-buffered saline; PSA: Prostate-specific antigen.

Table 3. Chip-based lysis methods.
Technology	Sample	Lysis volume	Time required	Result	Advantages	Disadvantages	Ref.
Heat-supported chemical and enzymatic lysis	500 µl UTM swab extract spiked with 10¹⁰ copies of in vitro transcript of an RSVB fragment	3 ml	10 min	Similar to conventional method	Easy, efficient, low integration effort	No time reduction and high power consumption due to high volume	[18]
Heat-supported mechanical/chemical/enzymatic lysis, variable concentration of specific detergents, monolith	100 µl whole blood spiked with 10¹–10⁶ cfu Gram-negative (Escherichia coli) and Gram-positive (Bacillus subtilis and Enterococcus faecalis) bacteria	450 µl	Not given	LOD: 10² cfu E. coli/ml and 10³–10⁴ cfu Gram- positive bacteria/ml	Combined lysis and DNA isolation	Efficiency highly dependent on target cell owing to different visco-elasticity of cell walls	[32]
Laser-irradiated magnetic bead system	100 µl blood-spiked HBV or E. coli at a volume ratio of 1:2, final HBV/E. coli concentration = 10⁴–10⁶ copies/ml	10 µl	3 min	~90% capture efficiency	Reduction of binding time from 20 to 3 min does not affect lysis efficiency significantly	Larger concentration of magnetic beads needed for HBV isolation, volume limited (CD)	[109]
Shear and strain induced forces by flow through a polymer monolith	50 µl, 10¹–10⁵ cfu E. coli/ml + 1% whole blood in hematuric urine	100 µl	30–40 min (inclusive DNA isolation)	Down to 10¹ cfu/ml detected	Combined lysis and DNA isolation	Will not work for all organisms due to the DNA binding conditions needed	[110]

CD: Centrifugal disk; cfu: Colony-forming unit; HBV: Hepatitis B virus; LOD: Limit of detection; RSVB: Respiratory syncytial virus B; UTM: Copan's Universal Transfer Medium.

Table 4. Various chip-based technologies for the extraction and purification of nucleic acids.
Technology	Sample	Time required	Result	Advantages	Disadvantages	Ref.
Chitosan matrix	20 µl nasal swab eluate containing 200 pg–2 ng of influenza A viral RNA	~15 min (inclusive of lysis)	200 pg viral RNA in nasal swab eluate containing 20-fold excess (mass) of cellular nucleic acids	Aqueous extraction method avoids reagent-based PCR inhibitory compounds	High elution volume, manual aliquoting of RNA eluate	[64]
Silica-impregnated polymer monolithic columns	50 µl, 10¹–10⁵ cfu Escherichia coli/ml + 1% whole blood in hematuric urine	30–40 min (inclusive of lysis)	Down to 10¹ cfu/ml detected	Combined lysis and DNA isolation	Will not work for all organisms due to DNA binding conditions needed	[110]
Magnetic bead-based DNA extraction and purification	25 µl whole blood, 10⁵ cells (E. coli and Bacillus subtilis)/5.4 µl	~15 min	Comparable to the standard bench-top extraction process	Combination of volume reduction and target isolation	~10% loss of beads, interference of magnetic beads within eluates with A260 readings possible	[114]
Combination of a C18 reversed-phase column with a monolithic SPE	125 µl of 0.08 µl blood/µl lysate sample	Not given	10 µl whole blood, 99% of DNA traverses the C18 phase 70% of protein retained, 69 ± 1% DNA extraction efficiency	Protein capture before DNA binding leads to ~100 times increase in DNA extraction, 2-propanol wash step eliminated	High monolith derivatization time (75 min) needed to provide sufficient binding capacity, long equilibration times required	[115]
Combination of silica-based SPE with ion exchange SPE (chitosan)	500 µl 1:125 diluted whole blood or semen or 40 µl 1:500 diluted whole blood	~1 h	In contrast to conventional micro-SPE, full profile (seven loci) can be detected	Silica matrix for volume reduction (50- and 14-fold for DNA and RNA purification, respectively) and chitosan for elimination of PCR inhibitors	Reduction of free DNA binding sites in more complex samples due to unspecific protein binding	[116]
Isotacho-phoresis with fluorescent labeled molecular beacons	2 µl bacterial lysate, 10⁸ cfu/ml E. coli or resuspended pellet from human urine	15 min (complete analysis)	10⁶–10⁸ cfu/ml detected	Increasing the total amount of focused sample by continuously focusing new sample, quantification possible	Complex samples will not result in pure target fractions	[117]
Free-flow magnetophoresis and laminar-flow profile	250 µl E. coli lysate, 1.84 × 10⁹ cfu/ml	2 min	147 ± 28% DNA recovery compared with standard protocol	Simple fluidic channel network, continuous extraction	High space required, need for proper controlled fluid conditions	[119]

cfu: Colony-forming unit; SPE: Solid-phase extraction.

Table 5. Integrated systems dedicated to the identification of pathogens.
Study (year)	Sample	Lysis	DNA/RNA extraction	Amplification	Detection	Reagent storage	Time required	Result	Chip material	Actuation	Advantages	Disadvantages	Ref.
Sauer-Budge et al. (2009)	50–400 µl, 1.25 × 10⁶, 6.25 × 10⁶, or 12.5 × 10⁶ Bacillus subtilis cells	Chemical/mechanical lysis (shear and strain induced forces)	Porous polymer monolith with embedded silica particles, chaotropic salt	PCR, Taqman® assay, ceramic heater and air cooling	End point fluorescence detection accomplished with an optical spectrometer	Tanks off chip	Not given	1.25 × 10⁶ B. subtilis cells detected	Zeonex® 690R, CNC machined structures (compatible with injection molding)	Three syringe pumps, pneumatic dispensers valveless chip, remote valve-switching technique	Simple design, no active components on-chip	Bacterial lysis not completely efficient, monolith causes variations in flow resistance from chip to chip, variability in output signals	[7]
Ritzi-Lehnert et al. (2011)	500 µl UTM swab extract, spiked with 10¹⁰ copies of in vitro transcript of an RSVB fragment or an ADV standard	3 ml, 10 min, heat supported chemical and enzymatic lysis	Binding of nucleic acids to magnetic silica beads during lysis, chaotropic salt, elution in 100 µl water	120 µl, 30 min, nested RT-PCR, multiplex, thermal cycling by movable heating brackets at constant temperatures (double-sided heating, PCR carousel)	Nonintegrated, labeling with color-coded QIAGEN LiquiChip beads on-chip	PCR reagents freeze-dried on-chip, liquid buffers in reagent cartridge for 24 runs	<1 h	50–81% of the signal intensity of the references	Injection-molded PC chip, bonded with PC foil	Syringe pumps, turning valves inclusive metering structures, magnetic stirring	Currently, the fasted high-volume PCR on-chip, on-chip nested PCR leads to enhanced sensitivity, contamination risk of nested PCR eliminated due to closed system	Complex operating device, swab extraction and detection not integrated	[18]
Easley et al. (2006)	6 µl whole blood of Bacillus anthracis-infected mice or 8 µl nasal aspirate of Bordetella pertussis-infected patients in 100 µl lysis buffer	Not integrated	750 nl of lysed sample, SPE, chaotropic salt	550 nl PCR, 10 min, IR thermal cycling	Electrophoretic amplicon separation, performed under high fields with laser-induced fluorescence detection	None	<30 min	B. anthracis in 750 nl of whole blood, B. pertussis in 1 µl nasal aspirate (patient)	Two structured glass layers (etched), one poly(dimethyl siloxane) membrane, one unpatterned glass layer	Syringe pump, elastomeric membrane valving	Combination of different technologies for effective analysis, fast temperature cycling (no additional thermal mass)	Referred as the first, fully integrated lab-on-a-chip system although lysis is off-chip, manual loading of electrophoresis matrix during PCR	[29]
Chen et al. (2009)	50 µl, ~10⁶ Bacillus cereus/ml	50 µl, two-step heat supported enzymatic (30 min) and enzymatic/chemical (10 min) lysis	Porous silica membrane, chaotropic salt, several 50 µl elution fractions	5 µl DNA eluate in 10 µl PCR, 30 min, thermal cycling by a thermoelectric element	15 min labeling with avidin-conjugated, upconverting phosphor particles, lateral flow strip and laser scanner	None	~1h	<10 DNA copies with the IR-PCR system	Thermally bonded three-layer laminate of PC sheets, CNC machined structures	Syringe pumps, thermoelectric phase-change 'ice' valves	Simple detection strategy, no mechanical valves on chip	Very time-consuming lysis, PCR inhibitors in initial elution fraction	[93]
Lui et al. (2009); Chen et al. (2010); Qiu et al. (2011)	100 µl, 10³–10⁷ virions/ml (viral armored RNA HIV and HIV I virus), 1.4 × 10⁴–1.4 × 10⁷ B. cereus/ml in saliva samples	200 µl, 1 min for RNA viruses, 30 min for B. cereus, chemical lysis	Silica membrane (glass fibers), chaotropic salt, 25 µl elution	5 µl DNA/RNA eluate in 25 µl RT-PCR, 30 min, thermal cycling by a pair of thermoelectric elements (double-sided heating)	Labeling with digoxigenin and biotin, detection with upconverting phosphor particles on lateral flow strip and IR laser reader	Liquids (in flexible pouches) and dry PCR reagents (passivated with a paraffin film) prestored in the cassette	~1h	Detection limit: 10³–10⁴ B. cereus/ml, 10⁵ HIV virions/ml	PC, CNC machined structures (compatible with injection molding)	Vacuum pump, linear motors to actuate on-chip pouches and diaphragm valves, vibrator for mixing	Robust fluid handling by pouch actuation, simple detection strategy, no exchange of liquids between analyzer and cassette, a similar cassette can be used for real-time detection	Reagent drying at room temperature in moderate vacuum may adversely affected enzyme activity	[120–122]

ADV: Adenovirus; CNC: Computerized numerical control; IR: Infared; PC: Polycarbonate; RSVB: Respiratory syncytial virus B; RT-PCR: Reverse-transcription PCR; SPE: Solid-phase extraction; UTM: Copan's Universal Transfer Medium.

Box 1. Technical challenges during the development of lab-on-a-chip systems.
Sample collection/transfer Macro–micro transfer Assay Sample preparation Modification (selectivity/sensitivity) Metering, merging, mixing (integration) Surface coating: choice of material Technologies Temperature management (integration) Valve technology Actuation (fluid transport) Sensorics (sensitivity) Electronics (automation) Analysis of results Presentation of results: bioinformatics, user interface (automation) System requirements Reproducibility, robustness, shelf life, maximal output (parallelization and multiplexing), minimal size, minimal costs Sample preparation methods Filtration/extraction/isolation/concentration Centrifugation Electrophoresis Magnetic separation Mass spectroscopy Fluorescence-activated cell sorting Solid-phase extraction RNA/DNA isolation Lysis, hybridization Amplification Chromatography

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.