Medical Device Link.

MOLECULAR DIAGNOSTICS

Measuring bacterial growth, bacterial cell viability, and antibiotic resistance is a major objective for many IVD technologies, and an important factor in human healthcare and many industrial areas such as food production, water safety, and pharmaceuticals. Traditional culture techniques for such measurements still predominate and can be extremely sensitive (e.g., a detection limit potentially as low as 1 cfu). However, they are very slow, often with culture periods of more than 24 hours depending upon the microorganism.

A number of rapid methods have been developed to provide shorter analysis times in which there is a need for rapid results.¹ Such methods include nucleic acid testing (NAT) technologies based on specific genomic DNA and RNA amplification processes (such as polymerase chain reaction (PCR) and transcription mediated amplification (TMA), which can provide fast results within a few hours. Such molecular diagnostics techniques are ideally suited for detecting and identifying specific organisms or pathogenic species. However, they do not provide accurate information on the viability of the bacterial cells present in the sample.^2,3

One approach for adding this crucial functionality has been to focus instead on mRNA as the diagnostic target.⁴ mRNA has a short half-life and disappears rapidly from the cytoplasm upon cell death. Hence, mRNA content is an indicator of the cell’s viability. However, measuring mRNA for this application is not commonly used since reverse-transcriptase PCR is more complex and expensive than standard PCR processes. Moreover, mRNA can persist for long periods in dead cells, limiting the general usefulness of this technique.⁴

The alternative and most widely used nonmolecular diagnostic technique to determine cell viability is the measurement of the intracellular adenosine triphosphate (ATP) content. Such techniques employ luminescence generated by the firefly luciferase reaction and provide a quantitative measurement.^5–7 While ATP determination by luminescence is rapid, sensitive, and simple to perform, one disadvantage is that laboratories must invest in and maintain sophisticated luminescence equipment for what may be a single application.⁸

Detecting ATP

Iseao Technologies Ltd. (London) has developed a ligase-mediated ATP amplification assay (LiMA) that combines the measurement of ATP with a NAT format. This assay takes advantage of the significant improvements that have been made in NAT technologies, such as PCR and TMA, and other isothermal systems such as strand-displacement amplification (SDA). The aim was to achieve a performance comparable to the direct luminescence-based systems for ATP by substituting amplified double-stranded DNA synthesis for the photometric output signal generated by the luciferase reaction.

Figure 1. (click to enlarge) The LiMA principle.

Figure 1 illustrates the principle of the LiMA technology. DNA ligase, an ATP-requiring enzyme, is used to join two oligonucleotides in a nicked DNA substrate and create a template that can be amplified in a DNA amplification reaction.⁹ The ligase is treated with pyrophos­phate to de­adenylate the active site, leading to an enzyme that is inactive until an ATP molecule is bound that can recreate the covalent enzyme-AMP intermediate. This then leads to the generation of the intact phosphodiester backbone in the nicked substrate, forming a template for a subsequent DNA amplification process, typically PCR.

Figure 2. (click to enlarge) The conversion reagent.

The optimal approach for the most efficient use of an ATP molecule is to immobilize the deadenylated ligase and the nicked DNA substrate on the surface of a paramagnetic bead in order to provide a rate enhancement due to the proximity of the enzyme and substrate on the bead surface. This is the conversion reagent (see Figure 2). Advantages of this approach include the ability to use larger sample volumes in LiMA than the luciferase-based luminescence techniques.

Inhibitors of the NAT process that are often found to be present in the specimen can also be removed by washing the paramagnetic beads prior to amplification of the ligated substrate.^10,11 In this context, luminescent ATP assays can be affected by inhibitors of the firefly luciferase reaction, and these cannot easily be removed.¹² LiMA may have applications in situations where luciferase is sensitive to inhibitors in the sample, but ATP-dependent DNA ligase is not.

To generate the conversion reagent, a deadenylated DNA ligase by New England Biolabs (Ipswich, MA) and the nicked DNA substrate

5’GCCGATATCGGACAACGGCCGAACTGGGAAGGCGCACGGAGAGA3’, 5’CCACGAAGTACTAGCTGGCCGTTTGTCACCGACGCCTA3’, 5’TAGTACTTCGTGGTCTCTCCGTGC3’

were coupled to amine paramagnetic beads by Dynal AS (Oslo, Norway) using suberic acid hydroxyl succinate ester by Sigma Aldrich (Poole, UK), washed in standard phosphate buffered saline (PBS) using a magnet, and stored in PBS. This reagent is stable for more than six months at 2–8°C.

The following protocol was used for measuring ATP with the LiMA process:

• Add conversion reagent to sample (preferably in the range of 50 µl to 1 ml) and incubate for 15 minutes.
• Capture the conversion reagent using a magnet, and wash three times with PBS.
• Add the washed beads to a conventional PCR mastermix containing the double-stranded DNA intercalator dye Syber Green by Eurogentec (Seraing, Belgium).
• Proceed with real-time PCR over a 60-minute period. The Chromo4 real-time system by MJ Research Inc. (Waltham, MA) was used to measure fluorescence increase in microplate wells.

Table I. (click to enlarge) Comparison of LiMA performance with a commercially available luciferase product.

The response of the LiMA technology to ATP is linear over a wide range of ATP concentrations.¹³ Table I shows the current performance of LiMA versus a typical firefly luciferase reagent available commercially. At its current level of development, LiMA’s detection limit is approximately 10–100 times higher than the firefly luciferase luminescence reaction. However, since the sample volumes can be larger with LiMA, the practical sensitivity performance with real samples is comparable between the two techniques.

Detecting Bacteria

Figure 3. (click to enlarge) Standard curve for S. aureus. The function 2exp[CtO–CtATP] is referred to as the growth index, where CtO–CtATP is the difference in cycle number when the fluorescence signal generated in the PCR crosses a predetermined threshold level. Log growth index is plotted vs. log cell number.

Figure 3 shows a standard curve for log phase Staphylococcus aureus grown in a standard culture medium. Calculated by using 2.5× standard deviation on 10 replicates of the zero standard, LiMA’s detection limit ranges from 10⁴–10⁵ S. aureus cells in a sample volume of 1 ml of the bacterial culture. Using the accepted average bacterial ATP content of approximately one attomole per actively growing log-phase cell, this detection limit for bacterial cells corresponds to 10–100 fmole ATP in LiMA after cell lysis and is consistent with the detection limit for ATP (see Table I). This sensitivity performance in a sample volume of 1 ml is the same or better than the fire­fly luciferase luminescence method in bacterial cultures.^14,15

Determining Antibiotic Resistance

The following protocol was used for detecting antibiotic resistance and determining a clinically relevant minimum inhibitory concentration (MIC):

• Bacterial strains (S. aureus and P. aeruginosa) from the American Type Culture Collection with defined MICs were grown in the presence or absence of antibiotics for three hours.
• A 1-ml sample from the culture medium was lysed by adding 50 µl of 2 N NaOH containing 2% Triton X-100 and heating to 95°C for 3 minutes to release ATP, and then neutralized after 5 minutes incubation with 50 µl of 2.0 N HCl.
• A conversion reagent was added and incubated for 15 minutes.
• Conversion reagent beads were washed three times with PBS, and real-time PCR was carried out as described above.

Figure 4. (click to enlarge) Response of S. aureus to oxacillin (a) and MRSA (b).

Figure 4a shows that growth is inhibited in a sensitive strain of S. aureus by the presence of an antibiotic in the culture medium, while Figure 4b shows that the MRSA strain was resistant to the antibiotic as expected. Figure 5a shows the inhibition of growth of P. aeruginosa by gentamycin; the MIC was determined to be 0.5 µg/ml. Figure 5b shows the inhibition of growth of the same strain by ciprofloxacin; the MIC was 0.125 µg/ml. In both cases, the MIC determined by LiMA was within one growth doubling of the reported MIC for this strain.

Diagnostic Applications

Figure 5. (click to enlarge) Response of P. aeruginosa to gentamycin (a) and ciprofloxacin (b).

The LiMA technology has applications in which bacterial cell number, cell growth, and cell viability need to be determined. One of LiMA’s features is the ability to access or scavenge ATP from a range of sample vol­umes, even in the 1–100- ml range. This is an advantage for LiMA over firefly luciferase–based luminescence protocols that are limited to sample volumes of 50–100 µl since there is no inherent way to design an equivalent ATP capture approach in this system (see Table I).

This feature of the LiMA technology enhances sensitivity, which is useful in applications where bacteria are present in low numbers in large volumes of fluid, such as potable or pharmaceutical-grade water, wastewater systems, and beverages, and in manufactured items such as personal care products. The bacterial ATP in such samples can be accessed directly using the LiMA technology by lysis in situ, followed by ATP capture with a conversion reagent. There is no requirement to use complex filtration systems to capture and concentrate the bacteria first.¹⁶ The LiMA process is more cost-effective in the laboratory environment than the filtration-based luminescence products for this sample type.

In the LiMA process, enzyme inhibitors and fluorescence quenchers in the sample can be washed away prior to the subsequent NAT detection process, avoiding potential problems with the amplification reaction. The DNA ligase may also prove to be more resistant than luciferase to inhibitors in the sample. This allows bacterial ATP measurements to be made in sample types (e.g., certain foods, contaminated water, blood, and urine) that could not be used with the firefly luciferase–based luminescent technologies.

Another feature of LiMA is that the system’s dynamic range is higher than the firefly luciferase bioluminescence reaction. This can be an advantage when high numbers of bacterial cells are present, which could require sample dilution and re-test.

By developing an ATP measurement technology that works with conventional NAT chemistries, it is possible to combine genotypic tests for bacterial identification with phenotypic tests for viability and growth on the same instrumentation platform. The results presented in this article for determining MRSA confirm that a phenotypic antibiotic-susceptibility assay is feasible on a NAT platform.

The LiMA reagent kit is simple and cost-effective, containing the lysis reagents, the conversion reagent, and a wash buffer. The other required reagents are common NAT chemis­tries that are available from other suppliers. The total assay time for a LiMA analysis is approximately 90 minutes; the actual duration depends on the processing ability of the NAT platform. LiMA is best suited for a laboratory scenario in which there is preexisting NAT equipment, thereby avoiding further investment in luminometers or other specialist equipment. In addition, the LiMA procedure can be automated to achieve high throughput if desired.

Conclusion

It was the original intention to develop LiMA as a laboratory-only product. However, PCR lab-on-a-chip systems employing advanced microfluidics, onboard thermal cycling, and real-time/kinetic detection have developed rapidly. It is possible that a LiMA product could be developed for field use in hygiene or environmental testing (e.g., biodefense applications) as an alternative to the existing handheld firefly luciferase–based luminescence kits.16 For example, LiMA chips would be applicable in situations where high sample volumes need to be processed, environmental or industrial samples are inhibitory to firefly luciferase but not to ATP-dependent DNA ligase, or rapid near-patient assessment of a suspected bacterial infection needs to be done.

Stuart Wilson, is a founder and director at Iseao Technologies Ltd. (London, UK). He can be reached at stuart.wilson@iseao.co.uk.	Sharon Banin, PhD, is senior scientist at Iseao Technologies Ltd. (London, UK). She can be reached at sharon.banin@iseao.co.uk.	Christopher Stanley, PhD, is a founder and director at Iseao Technologies Ltd. (London, UK). He can be reached at chris.stanley@iseao.co.uk.

 

---