GB2415254A - Isolation and purification of biochemicals - Google Patents

Abstract

A method of selecting process steps for the separation of components, comprises the steps of providing a database of physical properties of potential components and of potential process steps, selecting from the potential components a plurality of actual components and extracting physical properties of the actual components from the database, displaying the actual components in graphical form in at least two dimensions, each dimension corresponding to a physical property, selecting from the potential process steps a candidate process step; and displaying in the graphical form the separation effect on the actual components of the candidate step. The application also discloses a corresponding computer based method of selecting process steps for the separation of components.

Description

f' Arts Isolation and purification of biochemicals

Field of the invention

The present invention belongs in the field of Bioinformatics.

Background to the invention

Bioinformatics is the use of computer technology to store, organise, generate, retrieve, analyse and share genomic, biological and chemical information.

Recently the Human Genome Project has produced a plethora of genomic, biological and chemical data. The project identified thousands of target proteins that could have a beneficial effect in treating humans. The gene sequences databases and related information help scientists to determine whether and how a particular molecule is directly involved in a disease process. This in turn helps to identify better drug targets.

The manufacturing of such target drugs can be summarised in three steps: identify a target biochemical, e.g. a protein, which has the potential to be a drug; grow the target protein in a biological host, e.g. yeast, bacteria, milk, plasma, etc and finally; separate the target from the host.

A problem with present methods of analysing biochemicals such as proteins is that they are time consuming and expensive, and suffer from inefficiencies in detection, imaging, purification and analysis. Many proteins share the same biophysical properties, so it is usually difficult to clearly identify a property of the target protein which differs from the full complement of other proteins present in the host. For example two proteins may share the same characteristics of separation with temperature, which when plotted on a graph of temperature against time will show these two proteins occupying the same position in space and time. As a result most separation and purification methods involve a trial and error analysis, which results in long delays (usually years) before a drug can be brought to market.

It is the aim of the present invention to provide an improved and time saving tool which leads to the intelligent isolation of a product.

Summary of the invention

The present invention provides a method of selecting process steps for the separation of components, comprising the steps of; providing a database of physical properties of potential components and of potential process steps; selecting from the potential components a plurality of actual components and extracting physical properties of the actual components from the database; displaying the actual components in graphical form in at least two dimensions, each dimension corresponding to a physical property; selecting from the potential process steps a candidate process step; and displaying in the graphical form the separation effect on the actual components of the candidate step.

It also provides a computer based method of selecting process steps for the separation of components, that comprises the following steps: providing a database of physical properties of potential components and of potential process steps; selecting from the potential components a plurality of actual components and extracting physical properties of the actual components from the database; displaying the actual components in graphical form in at least two dimensions, each dimension corresponding to a physical property; selecting from the potential process steps a candidate process step; and displaying in the graphical form the separation effect on the actual components of the candidate step.

Brief description of the drawings

Figure 1 is a flow diagram of operations that are performed on a mixture of different proteins according to a particular embodiment of the present invention.

Figure 2a is a flow diagram of operations to specify the target and the matrix within which the target protein is found.

Figure 2b is a flow diagram of operations to selecting a process route.

Figure 2c is a flow diagram of operations to generate the process documentation.

Figure 3 is a flow diagram of operations of an example process of purification of Alcohol Dehydrogenase (ADH).

Figure 4 shows the standard route used in the process of ADH purification.

Figures 5a to 59 are a visualization of the matrix of properties in relation to the entire complement of proteins in the feed material, as each purification and separation process is carried out.

Figure 6 shows a simulation of gel processes.

Figure 7 shows an alternate route which results in better purification of ADH.

Figures 8a to 8i are a visualization of the matrix of properties in relation to the entire complement of proteins in the feed material, as each purification and separation process is carried out.

Figure 9 shows a simulation of gel processes which demonstrate the purification level at each stage.

Figure 10 is a flow diagram illustrating a second example of purification of a target protein: Lactoferrin from milk whey.

Figure 11 is a flow chart detailing the process steps in an example process route.

Figures 12a to 12i are a visualization of the matrix of properties in relation to the entire complement of proteins in the feed material, as each purification and separation process is carried out.

Figure 13 shows a simulation of gel processes to estimate the product purity and concentration at each stage of the process route.

Figure 14 is a flow chart detailing the process steps in an example process route.

Figures 15a to 159 are a visualization of the matrix of properties in relation to the entire complement of proteins in the feed material, as each purification and separation process is carried out.

Figure 16 shows a simulation of gel processes which demonstrate the purification level at each stage.

Detailed description of the drawings

The present invention encapsulates experimental data, theoretical models and heuristic information into a three-dimensional user interface to guide the user towards a bioprocess route. It links proteomic information with big-process technology.

Figure 1 shows a flow diagram of operations that are performed on a mixture of different proteins according to a particular embodiment of the present invention.

A three-dimensional graphical interface provides a visualization of the whole matrix properties compared to the product properties to identify the best biophysical properties to separate against. These properties may include density, charge or water resistance amongst others.

Potential routes to purification are identified based on process stream properties and desired scale of processing.

The next stage simulates one-dimensional and two-dimensional gel processes to indicate product purity.

Simulations of a range of processing operations (e.g. precipitation, chromatography, filtration etc) allows for the estimation of purification and yield factors for any chosen product in the matrix.

The protein properties are calculated based on amino acid sequences and any available experimental data for use in simulations and visualizations.

The above process steps are brought together to form a properties database, which allow for simulations and visualizations based on either experimentally measured values or theoretically calculated values (or a combination of both) of protein properties.

The central process allows the user to update the protein database by the addition of new proteins or proprietary experimental data on protein properties.

At this stage data on other biophysical entities (i.e. not just proteins) may be added, to use in simulations and visualizations.

After a process route has been selected a facility may be provided to automatically generate process documentation to guide laboratory, pilot or process scale protocols for implementing the purification (a process flow diagram may also be included as desired).

The process may also allow for the process stream properties (e.g. pH, ionic strength and temperature) to be changed at any stage and immediately effect displayed protein properties and subsequent simulations and visualizations.

The system may also allow one to step back and forward though the process route displaying protein and process properties and product purity and yield.

Another option is to allow the removal and insertion of unit operations at any stage with subsequent recalculation of the whole process route.

The provision of a connection to the major proteomics databases allows one to draw on globally available information.

Figure 2a is a detailed flow diagram of operations to specify the target and the matrix within which the target protein is found. Initially the product and the matrix within which the product exists or is expressed are specified. Additionally the scale and type of processing are defined, i.e. the volume and concentration of the material to be processed. This information influences the range of processing options which are sensible to apply. Further information may be required as to the properties and stability of the product protein.

Figure 2b shows a detailed flow diagram of process steps to selecting a process route. Here the user can select a variety of protein properties to display via the three-dimensional interface. The various characteristics of a target molecule or drug are plotted in a matrix of properties such as density, charge or water resistance in relation to the entire complement of proteins in the feed material.

This enables a very complex problem to be viewed simply and processing steps (unit operations) to be effectively selected to separate the target drug from its host material. The combination of the three-dimensional interface and heuristics provides a powerful tool to guide the user to a choice from the many possible process routes.

Figure 2c illustrates the process documentation generation steps. After a process has been selected, the document production process is eased by automatic generation of process flow diagrams and experimental protocols for the selected process solution.

The process will now be discussed in relation to two different specific examples, ADH purification from clarified yeast and purification of Lactoferrin from milk whey.

Example - ADH Purification from clarified yeast The first example demonstrates a simulation of purification of the target protein alcohol dehydrogenase (ADH) from a clarified yeast homogenate.

The example feed matrix is much simplified in this example (to more clearly demonstrate the process) and just 60 proteins are represented where typically several thousand proteins would be present.

There are many possible routes to purification, with different options at each separation stage - this is illustrated in Figure 3. As a first step, the clarified yeast homogenate may undergo a Precipitation Process or Ultra-filtration. Either of these routes may then be followed by further Ultra-filtration, Hydrophobic Interaction Chromatography, Anionic Exchange Chromatography or Cationic Exchange Chromatography, to name but a few of the available separation techniques. Any or all of these separation steps may be followed by further purification processes, e.g. Reverse Phase Chromatography, Size Exclusion Chromatography or Hydrophobic Interaction Chromatography.

The general rule is to Capture, followed by Intermediate Purification, followed by Polishing. The question left to the scientist is which unit operation and process conditions should be used at each of these stages from the many options available.

An experimental route, as mentioned above, is devised through a trial and error method. However, the present invention allows one to quickly visualise where the target protein lies in the complex protein matrix and to virtually fractionate and step forwards and backwards through the process to assess the performance of each separation stage and the combination of stages. A virtual gel simulates how a gel would look at any stage in the process to indicate the purification level.

A standard route to the purification of ADH is shown in Figure 4. The clarified yeast homogenate is first subjected to Ultra-filtration, the aim of which is to concentrate the protein solution prior to chromatography. There will be some level of purification based on size at this point. The cut-off is chosen so that no ADH passes through the membrane. This step is illustrated in detail in Figure 5b, using the present invention.

Figures 5c and 5d show in detail the Anionic Exchange Chromatography process.

The pH of the feed is shifted so that it is above the isoelectric point (IEP) of ADH, in this example by 0.5 - pH units. The ADH will be negatively charged and hence bind to an anionic resin. All proteins which are positively charged pass through the column. A gradient elusion profile with a lower pH solution allows ADH to be released from the column.

The final gel process is shown in Figures 5f and 59, before the gel process has occurred, figure 5f and after gel processing, figure 59. Here the polishing step is a size-based separation (gel process) which also increases the concentration of the solution. Figure 6 further illustrates a simulated gel process which demonstrates the purification level at each stage of the above route.

Figures 7 and 8a to 8i demonstrate an alternative route which may result in better purification of the ADH.

This example route uses two precipitation steps, the first with the ADH in the solid phase, the second with the ADH in the solution phase. The aim of the first precipitation step is to precipitate out the ADH for purification and simultaneous concentration. This is detailed in Figures 8b and 8c.

The aim of the second precipitation step is to keep the ADH in solution and further purify the ADH with a second cut, removing contaminants in the solid phase. This is illustrated in Figures ad and Be.

Hydrophobic Interaction Chromatography is a useful next step to take advantage of ADH's hydrophobic characteristics. This yields a high purification factor, as shown in Figures Of and 89.

The last step is, as in the standard route, polishing with gel chromatography, which leads to increased concentration, illustrated in the feed matrix diagrams in Figures 8h and 8i.

Figure 9 demonstrates the use of simulated gel processing to indicate the purification levels at each stage.

Example 2 - Purification of Lactoferrin from milk whey The second example demonstrates the simulation of purification of the target protein, Lactoferrin from milk whey. For the sake of clarity just 18 proteins are used in the feed matrix, typically the target protein would need to be separated from a mixture of several hundred different milk proteins. As in Example 1, there are many possible routes to the purification and separation of the target protein, Lactoferrin. This is shown in Figure 10, the steps may include any or all of the following processes: Precipitation Process or Ultra-filtration, Hydrophobic Interaction Chromatography, Anionic Exchange Chromatography, Cationic Exchange Chromatography, Reverse Phase Chromatography, Size Exclusion Chromatography or Hydrophobic Interaction Chromatography, to name but a few of the processes available.

The standard route to the separation of Lactoferrin protein is shown in Figure 11, which includes the steps of two precipitation steps, the first to precipitate out the Lactoferrin for purification and simultaneous concentration, the second to keep the Lactoferrin in solution and further purify the target protein with a second cut.

These precipitation steps are illustrated in Figures 12b to 12e.

Figures 12f and 129 show the Hydrophobic Interaction Chromatography phase, followed by the polishing gel process, shown in Figures 12h and 12i. Figure 13 shows a simulation of gel processes to estimate the product purity and concentration at each stage of the process route.

A preferred route of purification of Lactoferrin from milk whey is show in Figure 14 and includes the steps of Ultra-Filtration, to concentrate the protein solution prior to chromatography (see Figure 15b), followed by Cationic Exchange Chromatography. At neutral pH, Lactoferrin is one of a few positively charged species, it will thus bind to an cationic resin. All proteins which are negatively charged will pass through the column. A gradient elusion profile with a higher pH solution allows Lactoferrin to be released from the column. This is illustrated in Figures 15c and 15d.

The final step of gel chromatography is a size-based separation and should also increase the concentration levels, Figures 15f and 159. Simulated gel processes are shown in Figure 16, to estimate the product purity and concentration at each stage.

It will be appreciated that, notwithstanding the fact that the above examples relate to the selection of specific processes for separation from yeast and milk hosts, the invention is of general applicability in that it provides a selection methodology that is applicable to any combination of processes and hosts.

Thus, other processes can be included within the framework that the invention offers, including other known processes and processes that are developed in the future. Likewise, the host can, in principle, be any host including but not limited to microbial hosts such as bacteria, yeasts, moulds, algae, protozoa etc. mammalian cells such as hamster, insect, human etc. transgenic hosts based on plant, milk, egg, blood etc. or other forms of host such as plasma, milk or plant extracts.

Claims (8)

1. A method of selecting process steps for the separation of components, comprising the steps of; providing a database of physical properties of potential components and of potential process steps; selecting from the potential components a plurality of actual components and extracting physical properties of the actual components from the database; displaying the actual components in graphical form in at least two dimensions, each dimension corresponding to a physical property; selecting from the potential process steps a candidate process step; and displaying in the graphical form the separation effect on the actual components of the candidate step.

2. A method according to claim 1 wherein the database comprises biophysical properties of potential components based on experimental data.

3. A method according to claim 1 wherein the database comprises biophysical properties of potential components based on amino acid sequences.

4. A method according to claim 1 wherein the candidate processes include at least one process selected from the list comprising: Precipitation Process, Ultra-filtration, Hydrophobic Interaction Chromatography, Anionic Exchange Chromatography, Cationic Exchange Chromatography, Reverse Phase Chromatography, Size Exclusion Chromatography, Hydrophobic Interaction Chromatography, one-dimensional gel process and two dimensional gel process.

5. A method according to claim 1 wherein the display medium comprises a three-dimensional graphical interface for visualization of the biophysical properties.

6. A computer based method of selecting process steps for the separation of components, that comprises the following steps: providing a database of physical properties of potential components and of potential process steps; selecting from the potential components a plurality of actual components and extracting physical properties of the actual components from the database; displaying the actual components in graphical form in at least two dimensions, each dimension corresponding to a physical property; selecting from the potential process steps a candidate process step; and displaying in the graphical form the separation effect on the actual components of the candidate step.

7. A method according to any preceding claim further comprising the step of generating process documentation.

8. A method of selecting process steps for the separation of components substantially as hereinbefore described, with reference to the accompanying drawings.