Sleep apnoea diagnosis: serum proteins linked to disease severity

Sleep tight

A good night's sleep is often the key to a good day ahead but there are a number of people who are fooled into believing they have slept well after a night of regularly interrupted sleep. They suffer from the debilitating respiratory disorder known as obstructive sleep apnoea (OSA) which leads to excessive sleepiness during the day.

Strangely, people with OSA generally have no recollection of irregular sleep patterns due to the body's automatic response to them. The throat muscles become so relaxed that the airway narrows to cause snoring. Further narrowing closes the airway completely and breathing stops for around 10 seconds.

After this time, the brain reacts to the reduced oxygen supply in the blood and wakes up the sleeper who will start to take deep breaths. In most cases, the person will go straight back to sleep totally unaware of being awake.

The condition is usually diagnosed by overnight monitoring of the patient in hospital by polysomnography which can include recording of brain waves, muscle tone, chest and abdomen movements, oral air flow, heart rate, blood oxygen levels and sound and video recording. Clearly, this is an expensive process and alternative methods are being sought.

One approach is to search for serum biomarkers of OSA that could lead to a simple blood test and a transatlantic team of scientists has adopted the proteomic route. Bernabe Jurado-Gamez from the Reina Sofia University Hospital, Cordoba, Spain, with colleagues from the University of Cordoba, ABSciex, Madrid, and Comer Children's Hospital at the University of Chicago compared the protein levels of OSA sufferers and normal sleepers in a study published in the Journal of Sleep Research.

Serum signals

They gathered together a group of 30 male OSA sufferers with 10 each classified in mild, moderate and severe groups. In addition, 10 male control subjects were selected who displayed a low likelihood of OSA following clinical evaluation and polysomnography.

Blood samples were withdrawn in the morning after overnight polysomnography and depleted of the most abundant proteins using a commercial peptide ligand library.

The enriched low-abundant proteins were digested with trypsin and the resulting peptides were labelled with iTRAQ reagents for relative quantification of the peptides by mass spectrometry. The data were searched against a reference database for protein identification.

In all, 103 proteins were identified unambiguously and a comparison of their relative levels revealed some differences. A total of 30 proteins had different expression levels between controls and OSA sufferers, with 11 lower in OSA and 19 higher.

These results suggested that serum proteins might be useful as indicators of OSA, so the researchers took a closer look at the proteins and their relationship to the occurrence and severity of the disorder.

Proteins point to sleep apnoea

Among the down-regulated proteins, levels of thrombospondin and complement component 4-binding alpha were both reduced in all three OSA severity groups. In addition, the relative levels of vitronectin, clusterin isoform 2, pre-apolipoprotein E and antithrombin were lower in the mild and severe groups.

Among the 19 overexpressed group, 13, 7 and 5 proteins were raised in the mild, moderate and severe OSA groups, illustrating a relationship with disease severity. Among these, fibronectin, apolipoprotein D and an apoliprotein-B 100 variant increased with the severity of the condition.

Most of the differentially expressed proteins were associated with lipid and vascular metabolic pathways, which give an indication of the processes that are involved during OSA.

So, the contents of some proteins do change with the onset of OSA and others vary with the degree of the condition, suggesting that they might form the basis of a biomarker panel for diagnosing OSA from a serum test.

However, the researchers were at pains to point out the limitations of the study which would have to be removed before the results can be confirmed. A larger cohort covering both genders needs to be tested and a confirmed symptom-free, disease-free control group should be included for comparison. In addition, the effects of treatment on protein levels should be considered before any firm conclusion can be drawn.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.