GC×GC-TOFMS Volatome Profiling Reveals pH-Dependent VOC Signatures of Malassezia pachydermatis

Malassezia pachydermatis is a lipid-dependent skin yeast that can transition from a commensal organism to an opportunistic pathogen under specific microenvironmental conditions, including altered pH. To better understand chemical signaling associated with this transition, a joint study conducted by researchers at the University of Washington (Seattle, Washington) and the University of Wisconsin, Madison applied comprehensive two-dimensional gas chromatography–time-of-flight mass spectrometry (GC×GC-TOFMS) to characterize volatile organic compounds (VOCs) produced by M. pachydermatis cultured at three pH conditions.

The team analyzed 30 GC×GC-TOFMS chromatograms using supervised, untargeted tile-based Fisher ratio (F-ratio) chemometrics to identify class-distinguishing analytes. Postprocessing metrics (R and RSD) were introduced to disentangle yeast-derived VOCs from pH-driven media effects, and a support vector machine model validated analyte classification with high specificity and low error. This work represents the first GC×GC-TOFMS-based investigation of the pH-dependent volatome of M. pachydermatis, demonstrating the power of advanced multidimensional chromatography and chemometrics for resolving complex microbial VOC profiles.

LCGC International spoke to Lina Mikaliunaite, lead author of the resulting paper (1) about this research.

What role does pH play in influencing the commensal-to-pathogenic transition of Malassezia pachydermatis?

Malassezia is a common commensal microbe that can become an opportunistic yeast. Under specific microenvironmental conditions, it can cause dermatologic and systemic diseases in the host. pH has been suggested as one such factor that might trigger these changes in Malassezia. When the skin's pH shifts outside the normal healthy range of 4.0-5.5, it can lead to certain dermatologic and systemic conditions.

Why are lipid-rich environments particularly suitable for the growth of Malassezia species?

Malassezia species depend on lipids and thrive in sebaceous regions like the scalp, face, and behind the ears, where these yeasts can intake lipids from their host.

How might microbial volatile organic compounds (VOCs) contribute to communication and pathogenicity in Malassezia?

Understanding how microbes communicate through VOCs could be key to figuring out what makes yeast turn pathogenic. Many things influence the production of these microbial signals, like the type of substrate and temperature. Additionally, the mix of VOCs can change with pH levels. While some research is beginning to explore the metabolic pathways of Malassezia's VOCs at the genus level, unraveling the intricate volatome and identifying what triggers pathogenicity in Malassezia and other microbes remains a complex challenge.

What advantages does two-dimensional gas chromatography-time of flight mass spectrometry (GC×GC-TOFMS) offer over conventional one-dimensional GC–MS when analyzing complex biological samples like microbial VOCs?

GC×GC provides many advantages over one-dimensional chromatography when analyzing complex matrices. One key benefit is the ability to separate all analytes chromatographically before they are analyzed in the mass spectrometer. This reduces ion suppression and helps identify analytes that might otherwise co-elute. Additionally, using GC×GC can enhance the signal since the eluent from the first-dimension column is focused into a narrower peak, lowering the detection limit. This is especially useful with biological matrices, where analytes can be present at very low concentrations.

Why are chemometric tools essential for interpreting GC×GC-TOFMS datasets, and what challenges do they help address?

Chemometric tools simplify the analysis of these datasets, which is crucial because GC×GC chromatograms can be very complex, containing hundreds or even thousands of peaks. That's why these tools are so valuable for truly understanding GC×GC chromatograms. They include methods like the Fisher ratio (F-ratio) tile algorithm, which is excellent for analyzing different sample classes with multiple replicates. These analysis methods are now available in user-friendly programs like ChromCompare+ from SepSolve and ChromaTOF Tile from LECO, making GC×GC more accessible and easier for new users to explore.

Can you explain the concept of tile-based F-ratio analysis and how it aids in distinguishing sample classes?

The F-ratio measures the ratio of variance between different sample classes to the variance within the same class. This method involves dividing the chromatogram into small tiles of user-defined size, summing the signals within each tile, and calculating the F-ratio for each. The hitlist is then ranked based on the highest F-ratio values, with higher scores indicating analytes that better distinguish between classes. Although this workflow cannot directly classify samples, it helps identify potential analytes that differentiate classes. These analytes can then be used in predictive models, such as support vector machine (SVM), to develop classification algorithms.

What are the key differences between the R and RSD metrics introduced in this study, and what specific aspects of the data do they evaluate?

The R metric was designed to exclude the pH influence of the media and focus on the difference between media blanks and Malassezia. Meanwhile, the RSD metric aimed to identify the pH influence while minimizing the effect of differences between media blanks and Malassezia. By using these two metrics, we could better understand this complex dataset, which involved VOC differences between media and Malassezia and required an assessment of pH variations.

How were the three analyte categories (consumed, shared, produced) determined using the R metric, and what biological insights do these distinctions provide?

The R metric ranged from -1 to +1, and the cutoff for the three different categories (consumed, shared, and produced) was empirically defined as the point when analytes show less than a 2-fold difference between the sum of the media blank signals and the sum of the Malassezia signals. Consequently, this condition yields an R metric threshold of ±0.33. When this R threshold is applied, the three analyte signal pattern categories become clear: a high positive R indicates an analyte produced by Malassezia, a high negative R suggests that an analyte is consumed by Malassezia, and a small R (around 0) corresponds to an analyte with a similar signal pattern between the media blank and Malassezia.

What is the significance of achieving high true positive and true negative rates (TPR/TNR > 0.95) in the SVM model used for validation?

True positive rate (TPR) indicates the proportion of positive cases correctly identified, while true negative rate (TNR) shows the proportion of negative cases accurately classified. In this study, all three analyte signal pattern categories performed very well, with each achieving TPR and TNR of 0.95 or higher, where 1 is perfect. High values for both metrics suggest better model performance and fewer false positives and false negatives. This also confirms that our R metric approach was sound, and a blind SVM model assigned analytes to the same categories.

In what ways could this GC×GC-TOFMS-based volatome analysis contribute to understanding or preventing pathogenic outbreaks related to Malassezia pachydermatis?

To our knowledge, this is the first study analyzing the M. pachydermatis volatome using GC×GC-TOFMS, allowing a more detailed examination of the volatile compounds produced and consumed by the microbe. Microenvironmental changes, including pH, are thought to trigger the transition of M. pachydermatis from a harmless commensal to a pathogenic state. It is crucial to examine how the VOC profile changes with pH variations to better understand the metabolic processes of M. pachydermatis in different microenvironments.

Reference

1. Mikaliunaite, L.; Whitten, J. M.; Tokihiro, J. C. et al. Tile-Based Fisher Ratio Analysis with Support Vector Machine Regression Modeling of GC×GC-TOFMS Data of VOCs Produced by Malassezia pachydermatis Grown at Variable pHs. Anal. Chem. 2025, 97 (37), 20147-20155. DOI: 10.1021/acs.analchem.5c02699