Study At TSRI Links Dozens Of Proteins To Several Rare Muscle And Nerve Degeneration Diseases

Led by TSRI Professors Larry Gerace and John R. Yates III, the team used a technique called subtractive proteomics to identify 62 new proteins in the inner nuclear membrane of the human cell. The team demonstrated that 23 of these proteins are linked with strong probability to 14 rare muscle-wasting diseases such as congenital muscular dystrophy, Limb-Girdle muscular dystrophy, and spinal muscular atrophy, and several forms of the neurodegenerative Charcot-Marie-Tooth disease.

Knowing the proteins that may cause or contribute to these diseases is a first step in the long process of looking for ways to detect, prevent, or treat them.

This study has the potential to clarify a significant number of the more than 300 human dystrophies for which a causative gene has not been identified.

"To understand how these diseases happen, we need to understand more about the players--the network of interlinked proteins," says Gerace.

Muscular Dystrophies and the Nuclear Membrane

Many rare but devastating diseases have been linked to the inner nuclear membrane, which lines the nuclear envelope compartmentalizing the cell's genetic material or DNA. On the inner surface of the membrane is a structure referred to as the lamina. The lamina is important for maintaining the shape and size of the nucleus. It also contributes to the specialized functions of different human cells, for example, enabling muscle cells to perform their particular functions and brain cells to perform theirs.

The lamina is largely composed of proteins called lamins, which are like bricks that form a scaffold-like structure for the nucleus. The lamina also contains membrane proteins that dock at the lamins.

"There have been a number of human muscular and neuronal dystrophies that have been linked to [these] proteins," says Gerace. "When certain lamins and inner membrane proteins are mutated, they cause disease."

Because of this link between lamina proteins and disease, scientists would like to know the identity of all the proteins in the lamina, and previous studies have identified about 20 lamina proteins.

In their current study, Gerace and Yates used a technique called subtractive proteomics to identify 62 more candidate human nuclear membrane proteins.

In the study, TSRI Postdoctoral Fellow Eric Schirmer demonstrates that the genes encoding 23 of these candidate human nuclear membrane proteins are in regions of the genome that have already been implicated in 14 muscle- and neuro-degenerative diseases.

However, many of these regions have hundreds of genes in them; so the identification of these disease gene candidates should greatly focus identification of the culprits.

"It's highly likely that some of these diseases will be due to [the newly identified] nuclear envelope proteins," says Gerace. "This is a pretty big step forward."

The Power of Subtractive Proteomics

Where "genomics" maps the DNA sequence and seeks to identify all the genes in an organism, "proteomics" takes a step further by asking where and when those genes are actually expressed as proteins.

One of the most important techniques emerging for proteomics studies is humbly referred to as MudPIT--Multidimensional Protein Identification Technology-which Yates has pioneered in the last few years. Using this technique, scientists like Yates are able to analyze and identify an enormous number of proteins in a complex mixture.

MudPIT basically combines liquid chromatography (which is like a molecular "sieve" that separates a complex mixture into its component parts) with tandem mass spectrometry (which identifies the components based on their masses). The instrument detects these masses and uses sophisticated software to identify thousands of separate proteins.

But MudPIT alone was not enough in this case, because the inner nuclear membrane is in contact with other structures of the cell and cannot be isolated without contaminating material. Identifying which proteins are from the inner nuclear membrane and which are contaminants presented a huge problem.

So the team used a simple subtractive technique to deal with this. They analyzed the nuclear membrane components with contaminants, containing 2,071 different proteins, and subtracted out the separately isolated contaminants, which accounted for more than 40 percent of the membrane proteins. From this list they were able to apply computational methods to limit the final list of new human nuclear membrane proteins to 62.

The scientists then took eight of these proteins at random and demonstrated that they all indeed targeted to the nuclear membrane.

The new nuclear membrane proteins identified in this study map to chromosomal regions where the following dystrophies have been localized:

Congenital Ptosis, hereditary type 1

Charcot-Marie-Tooth Disease 2A

Congenital Muscular Dystrophy 1B

Limb-Girdle Muscular Dystrophy 2B

Charcot-Marie-Tooth Disease 2A

Facioscapulohumeral Muscular Dystrophy (FSH)

Spinal Muscular Atrophy, Types 1, 2, and 3

Limb-Girdle Muscular Dystrophy 1A/1BArthogryposis: neurogenic, mild

Blepharophimosis 2

Charcot-Marie-Tooth Disease 2A

Distal Arthrogryposis, type 2B

Congenital Fibrosis of Extraocular Muscles 1

Distal Myopathy

The article, "Nuclear Membrane Proteins With Potential Disease Links Found By Subtractive Proteomics" was authored by Eric C. Schirmer, Laurence Florens, Tinglu Guan, John R. Yates III, and Larry Gerace and will appear in the September 5, 2003 issue of the journal Science. See: http://www.sciencemag.org.

This work was supported by the National Institutes of Health.