GC-MS Analysis of Garlic Breath

Introduction

The characteristic breath smell associated with the ingestion of garlic is well known. The odour is derived from sulphurous compounds in the digestive tract which are absorbed into the blood and then exchanged with gases in the lungs, and hence exhaled (Cai et al., 1995). In 1989 Dr Isao Sakai patented a process which claims to produce a durable odourless garlic. The patent claims to inhibit a biochemical pathway within garlic which subsequently leads to a reduction in the level of mercaptans in the breath. The patent claims that it is the mercaptans in garlic breath which are the principal 'malodour ingredients in exhalations' and that eating products that have been made with 'Sakai' garlic will lead to sweeter smelling breath after the meal.

The technique of breath analysis has been used for well over a century primarily for medical and forensic applications. Application of the technique within the food industry is less well documented due to the problems associated with this type of analysis - sampling has been found to be dependent upon the rate of chewing, breathing and saliva flow. Natural variation when using human subjects (Elmore & Langley, 1996) can lead to inconsistency in results as conditions cannot be controlled as successfully as within a model system (Taylor, 1996). Differences in saliva, pH, metabolism and enzymes amongst individuals have been found to influence the way that the individual handles sulphur compounds (Mackay & Hussein, 1978). A good example of this is the association of a high saliva pH with lower concentrations of methyl and propyl mercaptan in the breath. The technique of breath analysis can be divided into two categories, those which require pre-concentration of the sample and those which do not (direct sampling).

Direct sampling is the more straight forward to perform, e.g. Miriami et al, (1989) analysed 5m1 of breath following the ingestion of garlic. However many volatiles are present at only low levels and without pre-concentration may go undetected. Injection volumes of more than 10 ml in a gas chromatographic system (CC) or a mere 1 ml in a gas chromatography mass spectrometry system (GC/MS) are unfeasible due to high back pressure problems within the system, peak broadening, loss of resolution and alteration in retention times (Mackay and Hussein 1978). An alternative approach to direct sampling is that of real time breath by breath analysis whereby the breath is directly introduced into a detector. The technique of atmospheric pressure ionisation mass spectrometry (API-MS) has been used by Benolt et aI, (1983) and Brauss et al, (1998).The breath from the nose of the subject is drawn directly into the ionisation chamber of a mass spectrometer, where the water vapour present acts as the chemical ionisation agent for the volatiles. The drawback to this system is that resolution is based upon mass only and as such compounds that produce ions with the same mass, e.g. stereoisomers and positional isomers cannot be differentiated (Brauss et al, (1998). Sensitivity of the system is dependent upon the type of volatile being analysed but the lower detection threshold is typically in the range 10 -100 ppb by volume.

The more popular approach is that of pre-concentration using an adsorbent trap. A measured amount of breath is drawn through a trap by pump assisted means. It the adsorbent of choice has a high affinity for water a pre column containing a hygroscopic mix can be employed to enhance the trapping ability by preventing saturation occurring and to prevent obstruction of the subsequent desorption stage (Philips and Greenberg, 1991). The packing of choice most often quoted is Tenax, a porous polymer which exhibits a low affinity for water (of which there are high concentrations within the breath) along with little selectivity or artefact formation. This technique has been used by Mackay and Hussein (1978) and Ruiz et al (1994) to collect volatiles produced upon eating raw onions and garlic. After drawing a measured amount of breath through the trap it is thermally desorbed onto GC system.

Volatile work is normally performed using a GC system utilising either a flame ionisation detector (FID) or the more sensitive mass detector. As the garlic contains many sulphur -containing compounds much of the work cited has employed an element specific detector such as the flame photometric detector, FPD, (Mackay and Hussein, 1978 and Ruiz et al, 1994), the pulsed flame photometric detector, P-FPD, the atomic emission detector, AED, (Cai et al, 1995 and Quimby et al, 1998) or the sulphur chemiluminescent detector, SCD. (Restek application note). This specificity aids both sensitivity and selectivity, for example the FPD is able to improve sensitivity by one or two orders of magnitude compared to an FID (Varian). The AED, P-FPD and SCD are all linear and eqimolar in response, unlike the FPD. Varian claim that the P-FPD improves sensitivity by two orders of magnitude compared to the traditional FPD and also matches the detectivity and selectivity of the more expensive chemiluminescence detectors.

The aim of this work was to investigate the levels of mercaptans quoted in the Sakai patent, namely methyl, ethyl and propyl mercaptan in the odourless garlic compared with a non treated variety (control). The experiment also focused upon three further compounds, allyl mercaptan, dimethyl disulphide and one chemical quoted in work performed upon the Sakai odourless garlic by the Sanko Chemical Institute Co. Ltd, allyl methyl sulphide. The first step of this process was to develop suitable methodology for a pre concentration technique. This included creating a sampling method which was acceptable to the subjects, a GC-MS method and a means of trap desorption, using the instrumentation available. Only once consistent chromatography with good resolution of the compounds of interest was achieved could the experiment commence.
Breath samples from subject one after ingestion of the control garlic bread were also directly injected into a sulphur specific P-FPD detector.