Ramping Up

Proteomics promised much when it was first conceived in the late 1990s as the protein equivalent of genomics. Although its initial hype and a lack of practicable realism dented early expectations, technology and expertise have surged forward in the last few years. Diverse applications have emerged from drug discovery to biosimilars, and biomarkers to sophisticated protein chemistry.

What is proteomics? It can be defined as "the analysis of proteins on an industrial scale." Protein chemistry has moved from a meticulous and relatively slow science to a highthroughput, full scale development in biology and the role played in it by proteins. In the process, proteomics has moved from the bench top to spawn an industry of its own, in fact more than one, and these have consequences for the pharmaceutical industry.

Growing Importance
With its maturation, proteomics has moved to play a complementary role to that of genomics. Deoxyribonucleic Acid (DNA) sequencing, gene mapping, transcriptomics, micro arrays - these are all critical techniques that serve to supplement the depth of biological data that is available in the protein world. Without genomics and the advances in computational capability under the heading "bioinformatics", proteomics would not have been likely to move out of its 1990s depression.

The latest developments in proteomics reflect the broad and growing number of techniques that come under the proteomics banner. This is because proteins are both the operational and functional molecules of life. In colloquial terms, DNA provides the recipe, but proteins are the meal.

The formal definition of proteomics is the "analysis of all the proteins in a cell, tissue or organism, or more simply, a system." Exemplifying the practical term of system, proteomics is used narrowly to study specific pathways within a cell, or broadly, with the example of the Brain proteome project centered in Germany, to map every protein that is involved in the operation of the most complex organ on the planet.

There is also the sub-proteome containing all the peptides in a system - the peptidome - a rapidly growing area of interest in the biotechnology industry. Proteomics has evolved to become more than just identifying a protein.

Most developments in proteomics have focused on a systems approach, and specifically system comparison and biomarker discovery. Applications range from human diseases such as cancer and diabetes, to salt tolerance and pathogen resistance in crops. This is achieved by comparative proteomics that use techniques such as chemical labeling directly into the mass spectrometer, or variations on traditional Two-Dimensional gel Electrophoresis (2DE) with fluorescent tags.

The first step in any system is to map the proteome and to obtain maximal protein coverage - and there are many ways to achieve this. The original mapping technique of 2DE is visual. It remains a cost-effective approach and is a technique that small laboratories can utilize.

Its weakness has always been reproducibility, with replicates of 3-5 gels being necessary to achieve quality results. These replicate gels are then combined into a master image, which has been made possible by software tools such as Progenesis, and these images are compared between samples. Proteins that are identified as changing in concentration with statistical significance can then be excised, (enzymatically digested into peptide fragments) and identified by mass spectrometry.

Key improvements in reproducibility have arisen with the use of fluorescent tags and plastic backed gels. The former, which can be exemplified by Differential in Gel Electrophoresis (DIGE), allows two samples that are labeled with different chromophores to be analyzed simultaneously on a single gel - providing savings in time, although specialist imaging equipment is required.

The fear that the precious gel could still tear and results be ruined, appears to have been overcome when GelCompany introduced durable and fluorescence compatible plastic backed gels. The remaining weakness of 2DE is low throughput coupled with its nonautomatable nature, which has led to the growth of comparative analysis directly by mass spectrometry.



Advancement of Technology
Instrumentation is the other key component to the reawakening of proteomics. The developments in mass spectrometry hardware have been impressive and it is difficult to find an instrument that will not serve a proteomics laboratory well. This has created a multi-million dollar industry that is illustrated by the US$1.1 billion acquisition of MDS Sciex/Applied Biosystems by Danaher.

The major manufacturers have enabled the sensitive mapping of many protein and peptide systems - 50 micrograms of protein extract from a cell culture can be sufficient to identify over 1000 proteins from that system, or a single scorpion sting can be mapped to identify over 300 new peptides.

The range of mass spectrometry instruments is vast, but is centered on two core ionization techniques - Matrix Assisted Laser Desorption Ionization (MALDI) and ESI (Electrospray Ionization), followed by a mass measurement device, popularly ion traps, and Time-Of-Flight (TOF).

Many instruments produce complementary protein identification results. The rule of thumb is that for a given system, MALDI will yield 50 percent of the proteins, and ESI 50 percent, with 25 percent common to each, while 25 percent of the proteins may never be seen. In any system, more sample will always improve the number of protein hits. However, the fundamental physico-chemical properties of proteins and peptides means that some molecules simply cannot be detected (eg, they do not ionize or are insoluble).

For any particular application, it is a matter of selecting high-throughput or precision, and innovative hybrid instruments which combine different principles that allow enough variation in mass spectrometer types to suit the needs of a proteomics facility.

As recently as ten years ago, it was standard practice to identify one protein in 24 hours with Edman N-terminal protein sequencing. It is now possible to sequence and reliably identify one protein in one minute with a mass spectrometer. This was the start of the first transformation of protein chemistry into proteomics.

High-Throughput Processing
Gigabytes of genetic data, sophisticated computing power, broader expertise and innovation have since propelled this forward. Built on the diversity of the mass spectrometers, a range of modular approaches have been combined to increase the efficiency and depth of proteome mapping.

In particular, High Performance Liquid Chromatography (HPLC) has taken over 2D gels as the method of choice for sample separation. Some of the most successful techniques are based on the Multidimensional Protein Identification Technology (MudPIT) approach, which uses a series of chromatography steps (usually ion exchange and reverse phase) to separate a complex mixture of protein fragments.

An essential step is that the entire protein sample is first digested into peptides, before it is applied to the HPLC. This ultimately presents a bioinformatics challenge, but saves time because the sample (ie, all the peptides) can be directly analyzed by mass spectrometry.

In order to compare the differential protein expression of two or more samples, the proteins from each source can be chemically and uniquely labeled before the separation experiment begins. Two popular techniques are iTRAQ, which can be used to compare eight tissue samples simultaneously, and SILAC, an isotopic method that is readily incorporated into cell culture media.

In these scenarios, the peptides are identified and simultaneously quantified, with the tags being used to indicate relative protein amounts. An alternative option to labeling is direct spectral counting whereby the intensity of each peptide ion is measured in the mass spectrometer, and this value is used for quantification. When successful, the comparative analysis produces new information on the key components of the system - biomarkers for that disease or trait.

The objective of current developments is to improve peptide separation and sensitivity, without producing so many fractions that the mass spectrometer is then occupied for weeks on end. The initial approach was ESI-MS, with the direct injection of HPLC eluates into the mass spectrometer.

This has the immediate advantage over the 2D gel approach where proteins are both separated and identified in the process. There is the contrast however, that while proteins in a 2D gel that is stained with Coomassie blue dye can be excised and analyzed months later,
HPLC protein fractions require prompt analysis because their stability is limited in this form.

Another development has been the advent of LC-MALDI mass spectrometry, whereby the HPLC eluates are mixed with matrix and are simultaneously spotted onto MALDI targets. This has a number of advantages: firstly, high-throughput TOF/TOF instruments can be used to process the samples; second, the samples are stable on the target and can be archived, to be processed at a convenient time; and third, the samples can be re-analyzed.

There is a degree of complementarity between LC-MALDI and ESI-MS, because each will identify different proteins, but the efficiency of LC-MALDI is giving it the impetus.

The coverage that is obtained by the different proteome mapping methods varies. However, there appears to be an emerging trend, which is reflected in the comparable analysis of heart systems that obtained approximately1000 (rabbit) proteins by LC-MALDI, approximately 650 (rat) by MudPIT LC/MS/MS and 375 (human) by 2D-gel electrophoresis. However, no one approach will tell the entire story.

Whichever system is used, vast amounts of data are created - one comparative proteome mapping experiment today can create a file that is nearly one gigabyte in size. This requires computing power and necessitates centralized super computers that can handle the data of many groups. The Australian Proteomics Computational Facility is one such example, which accepts data from any group in the Asia Pacific region.

Drug Discovery
The power of LC-MALDI and other related techniques for proteome mining also opens doors for the pharmaceutical industry - a means of discovering new drugs from natural products.

Using traditional screening techniques, venom has already been proven to be a source of peptide drugs with multi-million dollar blockbusters such as captopril (Brazilian Viper), exanatide (Gila Monster), and ziconotide (Cone Shell). By adapting sensitive proteome mapping techniques and combining bioinformatics tools, an entire venom such as that of the black scorpion can be analyzed, enabling new molecules to be identified, synthesized and their activities tested.



In particular, the peptidome fragment of the proteome is an obvious target, since it hits both the sweet spot of the mass spectrometer, and the (current) optimal deliverable size range of protein drugs. To emphasize this, the peptide market is growing twice as fast as overall pharmaceuticals, with a market size in 2007 that is estimated at over US$3 billion with a projected growth rate of more than 10 percent per year.

In the same year, there were 67 therapeutic peptides on the market, 150 in clinical phases, 400 in advanced preclinical phases, and globally over 100 pharmaceutical and biotech companies that were active in peptides. These were achieved with traditional approaches.

Encompassed in the study of single pathways is the need to fully understand the actions of each entity, and an essential component is Post Translational Modifications (PTMs) - with phosphorylation being the prime example. The use of specialist instruments such as Applied Biosystems' 4000Q hybrid ion-trap or Thermo's Orbitrap have made the dissection of signaling pathways a reality. Major pharmaceutical companies are spending several million dollars in this pursuit. Notably, cell signaling is not restricted to phosphorylation alone, and other mechanisms such as sulphonation and redox changes of thiol groups are gaining traction. Studies of such PTMs are unfolding using current proteomics technology.

A spin-off from this interest has been to open the door to all aspects of protein chemistry, and to provide a suite of tools for the emerging biosimilars market.

This represents one of the most understated elements of proteomics - utilizing its power as a precision approach for the quality control of biosimilars. This growing market is forecast to be worth between US$19 billion within five years according to MarketsandMarkets, and US$77 billion within 2 years, according to RNCOS. Biosimilars, also termed biogenerics or follow-on biologics, represent second generation versions of blockbuster biologic products. Biosimilars are expected to have a profound impact on all aspects of the pharmaceutical industry.

Monoclonal antibody type drugs are the biggest sellers, with demand for novel medications like insulin, beta interferon, G-CSF and coagulation factors growing considerably. The European Union and the US have some of the most advanced biosimilars sectors while Japan, Canada, Australia, and particularly India and China are upcoming significant players.

Complete molecular characterization, from the verification of amino acid sequences to disulphide bridging and the ratification of PEGylation sites is emerging. Many techniques remain in their infancy in traditional analytical labs, whilst accredited facilities are rare.

Agencies such as the US Food and Drug Administration (FDA) are starting to realize the power of this era of proteomics, and are pressing for reliable mass spectrometry-based data. Regional testing authorities such as National Association of Testing Authorities (NATA) in Australia, and Asia Pacific Laboratory Accreditation Cooperation (APLAC) are responding. It is probable that ISO/IEC 17025 laboratory standards will become an essential Quality Control (QC) requirement for advanced proteomics facilities.

Proteomics has become more than just a single protein identification method. With the engagement of the other 'omics platforms - genomics, transcriptomics, bioinformatics and metabolomics - today's scientists have a breadth of complementary techniques at their fingertips.