SPR determinations aid WFI purity

Water for pharmaceutical preparations must be carefully purified, especially if it is to be used in parenteral products administered directly into the body. Pyrogens, in particular, need to be excluded to ensure product safety

Pyrogens, or endotoxic lipopolysaccharides, also called endotoxins, are fragments of Gram-negative cell membrane released during normal bacterial cell metabolism. Also produced by the death of Gram-negative cells, they are powerful immune stimulants, and in the worst case scenario can cause Gram-negative sepsis. So it is essential that they are excluded during the manufacture of parenteral pharmaceutical products.

Checking that pyrogens are absent from purified water is not simple. Traditionally, it was done using rabbits, but introduced more recently, and nowadays much the most common testing method, is the LAL, or Limulus amoebocyte lysate test.

It is based on the observation of Bang in 1956 that when the horseshoe crab, Limulus polyphemus, is infected with Gram-negative organisms, it suffers a fatal intravascular coagulation. It was later shown by Bang and Levin that this reaction was produced specifically by the endotoxin's action on a clottable protein contained in the amoebocytes, which are the only circulating blood cells in the crab. Horseshoe crabs are an endangered species, living in the ocean off the north-east coast of the US. To harvest the blood, they are caught and bled, before being returned to the sea, They are marked to ensure they are not bled again. However, the crab population is dwindling as they are also used as bait by fishermen.

Bang and Levin prepared a cell lysate from washed amoebocytes that is extremely sensitive to the presence of endotoxin, and which could be used to determine the amount of pyrogen present in the sample. Essentially, if a sample of pyrogen-contaminated water is injected into a sample of lysate, it will clot, with the reaction being complete within an hour. The precise amount of pyrogen present can be determined by dilution techniques.

new approach

However, both these methods are very long-winded, and it is all too easy for further contamination to be introduced as the samples are transferred from the processing line to the laboratory for testing. As they are, by their very nature, batch processes, they are impossible to run continuously.

Infinitely preferable would be a continuous on-line monitoring method, but this would require a new approach.

Such an approach has now been developed by the Lorch Foundation and Cranfield University, which uses surface plasmon resonance (SPR) to measure the levels of pyrogens in water.

If water containing pyrogens passes the surface of a gold film coated with molecules that recognise and bind pyrogens, the higher the concentration of pyrogen in the water, the greater the increase in interfacial refractive index observed in the sensor.

This alters the optimal angle of incidence of light needed to create the SPR phenomenon, and can be monitored by measuring the change in angle. The change can be measured as a function of time, giving a dynamic picture of events at the sensor's surface.

Biacore manufactures a commercial SPR biosensor that operates on this principle for measuring real-time biomolecular interactions by monitoring changes in interfacial refractive index. This direct affinity sensing removes the need for labelled assay components to monitor biological interactions, commonly using fluorescence, enzyme or radioactivity labels, and offers the possibility of cumulative dose-type formats, which is essential if a process is to be operated continuously.

However, when such sensors have been applied in real-world applications, non-specific interactions between the components of typically complex samples and the sensor can all too often take place. This can severely affect the sensor's ability to detect changes accurately at low concentrations.

An answer to these problems would be an on-line direct cumulative dose affinity sensor for process control in high purity water systems. This would be able to monitor pyrogen levels without the need for frequent regeneration of the sensor.

The principle has been proved by work carried out by Dr David Cullen and his team at Cranfield University, who used a commercially-available SPR sensor and anti-lipopolysaccharide antibodies to detect endotoxic lipopolysaccharides at levels low enough to satisfy the pharmaceutical regulators.¹

A Biacore 3000 instrument was used to carry out the measurements. Two different mouse monoclonal antibodies raised against lipopolysaccharide were studied: an immunoglobulin G2a raised against a strain of E. coli, and cross-reactive with K. pneumoniae, Sh. sonnei and S. typhimurium from QED Biosciences; and an immunoglobulin M, raised against aKDO monosaccharide of the inner core of lipopolysaccharide from HyCult Biotechnology.

affinity recognition

As endotoxins from a wide range of different Gram negative organisms would need to be detected in a practical monitoring device, it was important to have a broad binding specificity. The endotoxin sources were lipopolysaccharide from a strain of E. coli, and Control Standard Endotoxin from Associates of Cape Cod.

Biacore CM5 sensor chips with gold surfaces and an immobilised layer of carboxymethyl dextran gel were chosen as the base sensor. These allowed standard protein immobilisation chemistry to be used, based on the covalent linkage of largely amine-containing amino acid side-chains in proteins to the gel's carboxyl groups.

Each of the sensor's four channels was labelled with one of the two affinity recognition layers. A new sensor chip was docked and primed in the Biacore instrument with a continuous flow of buffer at a flow rate of 5µl/min, and a succession of four different solutions were passed over the sensors in order to immobilise the antibodies onto them.

The surface refractive index was monitored throughout this process, and a plot of time against surface refractive index was used to determine the change in refractive index caused by the antibody immobilisation.

further studies

In order to carry out SPR analysis of lipopolysaccharide solutions, the Biacore instrument was first primed with phosphate-buffered saline. The SPR responses of antibody-labelled sensor channels were determined either singly or in sequentially-linked channel pairs, following repeated interspersed injections of buffer and various concentrations of one or other of the endotoxin sources. Readings were taken immediately before and after the injections, allowing the increases directly attributable to lipopolysaccharide binding to be determined from the difference.

Each sensor chip was used for repeated exposure to varying concentrations of the two endotoxin sources, with only buffer washes taking place between sample injections and no specific regeneration of the immobilised antibodies being carried out.

Comparing results for the two antibodies showed that the IgG is more sensitive and gives a linear response as endotoxin dose is increased. Both were able to detect lipopolysaccharides at a level of 0.1EU/ml ; well below the regulatory limit for pyrogens in purified water for injection, which is currently 0.25EU/ml.

Further studies have been carried out to demonstrate the potential of the cumulative dose sensor protocol and assess for how long the antibody-labelled sensor can be used continuously before it loses its activity.

Biacore's automated fluidics system was used to inject a sequence of samples that simulated a typical ultra-high purity water system — a low lipopolysaccharide loading interspersed with spikes of high concentration. Figure 1 summarises the results of a continuous 10h experiment. A single sequence of samples of 0.01, 0.05, 0.10, 0.50 and 0.01 EU/ml of control standard endotoxin was injected, interspersed with repeated blank samples. The first three samples were injected at even intervals over the first 90 minutes, followed by a continuous flow of blank sample before the remaining samples were injected during the final 90 minutes.

A linear increase in cumulative response was seen on the IgM coated sensor, and a concentration of 0.01EU/ml was routinely detected, with no apparent loss of antibody activity.

The Lorch Foundation is now looking for partners to commercialise the system, and patents have been applied for. Although the technique is still in the development stage, these results clearly show that, for the first time, a practical continuous on-line monitoring system for pyrogens in water is possible.

Problems with pure water systems could be pinpointed much more quickly if an on-line monitoring process were available to raise the alarm, saving the pharmaceutical industry vital time — and money.

Reference

1. J.A. Taylor, G. Barrett, W. Lorch and D.C. Cullen, Analytical Letters 2002, 35, 213

What Is SPR Technology?

Biacore International AB (Biacore) is the world leader in the development of surface plasmon resonance (SPR) technology for application in various areas of basic and applied science research such as biochemistry, protein engineering, proteomics and the discovery, design and evaluation of new therapeutic agents. The technology enables the continuous monitoring of biomolecular interactions, providing real-time kinetic information on molecular binding without the need for fluorescent tags, radioactive markers and frequently without the need for prior purification. This contrasts with many traditional "end-point" assays that only provide information on a single binding characteristic. Application of SPR technology provides detailed information about several parameters that characterize molecular binding events: - How specific is the binding between two molecules (specificity)? - How much of a given molecule is present and active (concentration)? - How fast does the binding proceed (kinetics)? - How strong is the binding (affinity)? SPR technology provides a number of key advantages over other methods of molecular analysis: - Real-time kinetic analysis of molecular binding interaction - giving an unparalleled level of valuable information on events as they occur; - Non-label technology - ensuring that biomolecules are not subject to alteration prior to investigation; - "Non-contact" detection in which the SPR measurement light does not enter the sample - thereby, eliminating the quenching or absorbance problems that beset all spectrophotometric and fluorescent techniques; - Minimum sample size and preparation required, saving both research costs and time; The technology enables the measurement of interactions between two or more biomolecules by immobilizing one type of molecule (the chosen "interactant") on the surface of a sensor chip" and passing a solution containing the other molecule(s) (the "analyte") over the surface under controlled flow conditions. The chosen interactant determines the specificity by which the analyte interacts. For example, a particular type of antibody may be chosen as the interactant and immobilized on the sensor chip surface to study its interaction with a target antigen. Similarly, a particular cell surface receptor may be immobilized to study its interaction with certain ligands. The sensor chip consists of a glass surface coated with a thin layer of gold that provides the physical conditions required for SPR. SPR is an electron charge density wave phenomenon that arises at the surface of a metallic film when light is reflected at the film under specific conditions. An evanescent wave extends beyond the sensor surface and detects mass changes on the surface. At no time in the measurement process does light enter the sample, thereby eliminating the quenching or absorbance problems that beset all spectrophotometric and fluorescent techniques. To prevent non-specific absorption of biomolecules onto the gold layer of the sensor surface, the gold is covered by a layer of "dextran" on which the chosen interactant is immobilized by the user. The user selects the chosen interactant and the method of immobilization, thereby, determining the specificity of the interaction being studied. The dextran layer is designed to provide a favourable environment for interactions between typical biomolecules and to minimise non-specific binding. Once the chosen interactant has been immobilized, a microfluidic system injects the analyte solution over the sensor surface, allowing binding to occur and enabling the quantitification of association rates. When the analyte solution completes its pass over the sensor surface, measurements can be made of the dissociation rates. Flow conditions are controlled through a liquid handling system made up of precision pumps and an integrated fluidic cartridge (IFC) that is responsible for both forming the flow cells on the sensor chip surface and for directing the flow of liquid over the surface. Over 2000 peer reviewed scientific papers have been published, SPR instrumentation in basic and applied research by leading university groups, national research centres and major pharmaceutical companies.