Automated Dual-Chamber Sampling System to Follow Dynamics of

Oct 15, 2018 - ... fructification of golden oyster mushroom (Pleurotus citrinopileatus), and microbial degradation of a food-related sample (Pacific w...
0 downloads 0 Views 992KB Size
Subscriber access provided by University of Sunderland

Technical Note

Automated Dual Chamber Sampling System to Follow Dynamics of Volatile Organic Compounds Emitted by Biological Specimens Cheng-Hao Chang, and Pawel L Urban Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b03511 • Publication Date (Web): 15 Oct 2018 Downloaded from http://pubs.acs.org on October 20, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Technical note (revision)

Automated Dual Chamber Sampling System to Follow Dynamics of Volatile Organic Compounds Emitted by Biological Specimens

Cheng-Hao Chang1,2, Pawel L. Urban1,3*

1

Department of Chemistry, National Tsing Hua University 101 Kuang-Fu Rd., Hsinchu, 300, Taiwan

2

Department of Applied Chemistry, National Chiao Tung University 1001 University Rd., Hsinchu, 300, Taiwan

3

Frontier Research Center on Fundamental and Applied Sciences of Matters, National Tsing Hua University 101, Sec 2, Kuang-Fu Rd., Hsinchu, 30013, Taiwan

* Corresponding author: P.L. Urban ([email protected])

ACS Paragon Plus Environment

1

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 26

ABSTRACT

Growth of microorganisms is often accompanied by the release of volatile organic compounds (VOCs). The released VOCs may qualitatively or quantitatively reflect the physiological states of microbial cultures. A number of VOCs are produced during microbial degradation of organic matter accompanying food spoilage. In order to characterize the dynamics of microbial VOC production, and enable time-dependent analysis, we have constructed a dualchamber sampling system, and coupled it on-line with mass spectrometry (MS). The biological specimen is placed in a gas-tight sampling chamber. A carrier gas is introduced to the chamber periodically to transfer the VOCs present in the specimen headspace to the atmospheric pressure chemical ionization interface of a triple quadrupole mass spectrometer. A control/blank spectrum is recorded before recording specimen spectrum at each time point, enabling signal comparison and subtraction. The custom-made electronic control unit— incorporating three Arduino microcontrollers—operates six pinch valves, a miniature air compressor, and triggers MS data acquisition. The automated dual chamber sampling system was first tested using standard mixtures to verify its analytical performance. To demonstrate the usefulness of this system in the studies of microbial volatomes, we implemented it in realtime monitoring of the growth of baker’s yeast (Saccharomyces cerevisiae), fructification of golden oyster mushroom (Pleurotus citrinopileatus), and microbial degradation of a foodrelated sample (Pacific white shrimp, Litopenaeus vannamei). The recorded VOC signals show characteristic temporal profiles with transient bands or plateaus corresponding to different stages of microbial growth and putrefaction processes.

ACS Paragon Plus Environment

2

Page 3 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

INTRODUCTION

Volatile organic compounds (VOCs) are produced by various biological systems – cells,1 tissues,2 and organisms.3 For example, ethanol vapors are released by microbial cell cultures during fermentation.4 Plant tissues produce a number of VOCs (e.g. terpenes), which fulfill specific ecophysiological roles.5 On the other hand, the VOCs present in human breath indicate certain physiological states including diseases.6 Some VOCs can be indicative of an infection with microbial pathogens.7 Volatile metabolites can be used as indicators of microbial growth.8 Some play important signaling roles for fungi in their natural environments.9 VOCs are released by organisms continuously or intermittently. They are also produced during microbial degradation of dead organic matter (putrefaction).10 Thus, the knowledge of temporal changes in VOC levels is relevant to different areas of applied research, including microbial physiology, mycology, ecophysiology, medicine, as well as food chemistry. Conventional biochemical protocols involve sampling real matrices, sample treatment, and chemical analysis.11 However, it is difficult to store samples containing VOCs for long periods of time due to the risk of analyte losses.12 Moreover, profiling dynamic biological systems requires collection of multiple samples at different time points. This kind of repetitive sampling, and subsequent sample storage, are tedious and susceptible to artefacts. Taking into account the limitations of the conventional sampling methodology, it is appealing to develop a real-time monitoring system combining biological experiments with modern analytical instruments. Such hyphenated systems should provide optimum conditions for the biological organisms of interest, and—at the same time—ensure efficient loss-free transfer of the released VOCs to an on-line detector. The detector—used for tracking microbial metabolites

ACS Paragon Plus Environment

3

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 26

in real time—should enable selective analysis of the VOCs emitted by the biological specimens. These requirements are fulfilled by mass spectrometry (MS). In fact, MS is ideal for real-time analysis13 with high sensitivity.14 Recent reports demonstrated the possibility of real-time MS monitoring of microbial and plant metabolism (e.g.4,15). Considering the limited stability of MS detection, the lack of blank spectra at every time point in time-course studies becomes problematic. Compared with typical UV-Vis spectrophotometers—accommodating two parallel light paths for referencing—MS instruments typically operate only one ion beam. Therefore, direct referencing is not normally feasible. Taking into account the long durations of experiments involving real-time analysis of VOCs, it is not possible to obtain blank spectra recurrently without introducing automated features to the experimental system. Our first goal was to develop a simple automated dual chamber system for real-time sampling and MS analysis of VOCs, which could simultaneously record specimen and blank/control spectra. The second goal was to implement the developed system in the monitoring of VOCs produced during growth of microorganisms and putrefaction. To achieve automated operation of the biological experiment and chemical analysis, we implemented inexpensive, easy-to-program, open-source electronic modules.16,17

EXPERIMENTAL SECTION

We have designed and constructed a prototype apparatus to integrate long-term biological experiments with MS (Figures 1A and 1B). Briefly, the specimen is placed in a gas-tight culture chamber (Figures 1C-1E). A carrier gas (nitrogen for yeast and shrimp analysis; compressed air for live mushroom analysis) is introduced to the chamber periodically to

ACS Paragon Plus Environment

4

Page 5 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

transfer volatile compounds from the headspace zone to the ion source of mass spectrometer. Operation of this system is controlled by a custom-made electronic control unit (Figure 1B). The specimen chambers were designed in SketchUp software (Trimble, Sunnyvale, CA, USA), and fabricated with acrylic plates (Chi Hwa Advertising, Zhubei, Taiwan; Figures S1A, S1B and S2; see the Supporting Information for design files). In each experiment, two chambers were used: one for the specimen and one for the control or blank. Moreover, two types of chambers were constructed for different specimens: (i) small (inner volume, 152 mL); and (ii) large (inner volume, 936 mL). The contact edges of the upper part (lid) were lined with rubber to eliminate leakage of carrier gas. The lid of the small chamber was fitted with three tubing ports: (1) to supply carrier gas; (2) to deliver fresh air; and (3) to transfer gaseous analytes to the detector (Figure S1A). The large chamber was also fitted with gas-transfer tubing but in a different configuration (Figure S1B). An auxiliary air flow supplied by a miniature air pump (WRF370CE-22170; Yingyidianji, Huizhou, China) passed through humidifying reservoirs filled with 300 mL of phosphate buffered saline18 solution (pH 7.4). Nitrogen and air were supplied to the small and large chambers, respectively, at a pressure of ~ 90 kPa. The sampling flow rates were 59.5±1.7, 57.4±2.0, and 14.6±1.4 mL min-1 in the case of nitrogen flow through small chamber, air flow through large chamber, and auxiliary air flow through small chamber, respectively. The system dead volume from specimen chamber outlet to APCI needle was ∼ 0.5 mL. A normally closed pinch valve (Valve 1; P/N 390NC12150; Asco, Florham Park, NJ, USA) was used to control the flow of the carrier gas. Another normally closed pinch valve (Valve 2; P/N 390NC12150; Asco) was used to control the flow of gas out of the chamber toward the ion source of mass spectrometer. A normally open pinch valve (Valve 3; P/N 390NO12150; Asco) was used to control the flow of air from the air pump to the specimen chamber. A parallel gas line—with an identical set of valves—

ACS Paragon Plus Environment

5

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 26

was installed at the control/blank chamber. Thus, the total number of valves was six. All the tests were carried out with the specimen chamber thermostatted at ~ 25 °C (incubator: LM420D; Yih-Der, New Taipei City, Taiwan). Safety glasses must be worn when handling compressed gases. The valves and the air pump were actuated with the aid of a relay board controlled by an Arduino 101 microcontroller (Intel, Santa Clara, CA, USA; Figure S3). An Adafruit Data Logger Shield (Ardafruit Industries, New York City, NY, USA), with real-time clock function, was coupled to an Arduino M0 PRO microcontroller board (https://www.arduino.cc/) to precisely control the time interval between each run of data acquisition. Additionally, an Adafruit 1.8" color TFT shield (Ardafruit Industries)—coupled to an Arduino UNO R3 microcontroller board (https://www.arduino.cc/)—was used to display updates on the experiment progress. The three microcontroller boards were connected to one another to enable communication during experiments. The programs were written in C++ language (Arduino software, see the Supporting Information), and uploaded to the microcontrollers over USB. Table 1 lists the main steps in the analysis routine. The chambers were coupled via Valves 2 and 4 (via polytetrafluoroethylene (PTFE) tubing; 2× length: 18 cm; OD: 1.5 mm; ID: 1.0 mm; Supelco, Bellefonte, PA, USA) with a micromixer (YMC, Kyoto, Japan) fitted with KeyChem Y-type PTFE insert (YMC, Kyoto, Japan), and further downstream (via PTFE tubing; length: 40 cm; OD: 1.5 mm; ID: 1.0 mm; Supelco) with the atmospheric pressure chemical ionization (APCI) interface (Duis; Shimadzu, Tokyo, Japan) of a triple quadrupole mass spectrometer (LCMS-8030; Shimadzu). For MS settings, see the Supporting Information.

ACS Paragon Plus Environment

6

Page 7 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

RESULTS AND DISCUSSION

Characterization of the automated dual chamber sampling system The sampling system was characterized by constructing calibration curves for 6 volatile analytical standards, some of which are yeast metabolites: (1) ethyl acetate; (2) ethyl propionate; (3) ethyl butyrate; (4) ethyl hexanoate; (5) ethyl octanoate; and (6) diacetyl. Calibration was performed by injection of carrier gas flowing through small specimen chamber that held standard solutions (20 mL, inside Petri dish) mostly at five concentration levels. The calibration datasets were fitted with linear functions (Figure S4). The limits of detection (LODs) spanned from 3.20×10-6 to 1.42×10-5 M for ethyl esters (Table S1). The LOD of diacetyl was 4.12×10−5 M. While the above LOD values refer to the concentrations in the liquid solutions loaded to the specimen chamber, they can readily be converted to gasphase concentrations: 3.69×10-10 – 2.02×10-7 M (following a separate gas chromatographic determination; Figure S5, Table S2). The repeatabilities and reproducibilities were evaluated and expressed by relative standard deviation (RSD), which ranged from 2.7 to 13.3% (10 injections in one day) and 3.5 to 8.0% (3 injections in 5 different days), respectively (Table S3). The response time of the system was estimated based on the sampling flow rate and system dead volume as ∼ 0.5 s. The experimentally determined response time is longer: ∼ 7 s (from opening Valve 1 after 15 s of data recording to the maximum signal in Figure 2B, blue line). This discrepancy is due to the time required to stabilize the flow rate after opening Valve 1. Because prolonged passage of carrier gas through the specimen chamber could cause vapor dilution, short sampling times (45-280 s; cf. Table 1) were used. Because of the short response time, it was not necessary to purge the whole volume of headspace to obtain mass

ACS Paragon Plus Environment

7

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 26

spectra with VOC signals. In order to minimize perturbation to the biological specimen (by drying the specimen and removing VOCs from the headspace), we limited the time spent on purging chamber headspace during and after each measurement. It should be noted that blank mass spectra reveal a certain number of system peaks (m/z 37, 47, 61, 65, 75, 101, 107, 111, and 129; Figure 2C), which may be related to the laboratory atmosphere contaminants as well as plasticizers present in tubing or rubber (used to seal the chamber lids). For that reason, we subtracted baseline spectra (10 s, 20 datapoints) from the specimen spectra in most cases (Supporting Information). In fact, the dual chamber sampling system allows the user to notice the system peaks in the spectra corresponding to the control (blank) chamber at every time point of long-term experiment. We have also conducted additional tests to verify stability and lack of carryover (Figures S6 and S7). Note that the decrease of the signal in Figure S6 is due to evaporation of volatile analytes, what is expected considering the long duration of this experiment.

Initial tests with yeast cultures To demonstrate the usefulness of the automated dual chamber sampling system in the studies of microbial volatomes, we performed initial tests on baker’s yeast (Saccharomyces cerevisiae) cultures. Agar medium Petri dishes were placed inside the sample and control chambers (for details, see the Supporting Information). The Petri dish in the specimen chamber was inoculated with ∼ 1.9×106 yeast cells, while the Petri dish in the control/blank chamber was not inoculated. Because of the automated features—enabled by open-source electronic microcontrollers and C++ programming—the entire ∼ 55-h MS analysis proceeded seamlessly without any intervention of the experimenter. The mass spectra, recorded at different time points, reveal emission of several VOCs by yeast cells (Figure S8). A few of

ACS Paragon Plus Environment

8

Page 9 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

the signals in the spectra are due to contaminants (e.g. present in the system materials), while other signals are clearly related to the biological specimen because their intensities increased over time in the sample datasets but not in the blank datasets (Figure S9; 7 prominent features). Figure 3A shows the evolution of two yeast-related signals (m/z 87, 89). The final datapoints at different times were obtained by averaging ion currents during 10 s (20 raw datapoints; cf. Figure 2A). The obtained time profiles are dissimilar, for instance: m/z 87 (putatively identified as diacetyl – a known yeast metabolite19) reached a plateau within ∼ 9 h, while m/z 89 (identified as ethyl acetate, cf. Figure S10C) showed a growing trend till ∼ 36 h. Ethyl acetate can originate from different synthetic reactions, including esterification of ethanol with acetic acid and intracellular transesterification, e.g. involving ethanol and acetylCoA.20 The signal of acetic acid (m/z 59) initially increased, it peaked at ∼ 31 h, and it subsequently decreased (Figure S9). In the previous study, acetic acid signal increased until reaching plateau.4 However, the experimental design of the previous study was dissimilar (liquid culture, glucose added during culture, different number of cells at the start). The rapid increase of ethyl acetate signal follows depletion of acetic acid, pointing to the incidence of a (trans)esterification process. This result is comparable with the result obtained by Khomenko et al. using PTR-TOF-MS.21 Importantly, ethyl acetate and acetic acid signals were not observed in the blank measurement involving recurrent analyses of headspace vapors in the non-inoculated agar plate (Figures S9 and S11). Using the specimen chamber and control chamber MS data, one can readily verify whether a signal is due to a VOC released by the biological specimen or a contaminant. Here, Kolmogorov-Smirnov test was used to compare the data from the specimen chamber and control chamber at each time point (see the Supporting Information). Thus, with the temporal MS data available for the two chambers,

ACS Paragon Plus Environment

9

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 26

one can rapidly distinguish the system signals from the potentially interesting specimenrelated signals. In the second experiment, both Petri dishes—in the specimen chamber and the control chamber—were inoculated with the same numbers of yeast cells. This time, the medium in the specimen chamber contained 13C6-glucose as the main carbon source, while the medium in the control chamber contained

12

C6-glucose (same as the specimen chamber in the previous

experiment). As expected, the yeast cells grown on the

13

C6-glucose medium incorporated

carbon-13 into metabolites, what is revealed by the apparent “shifts” of peaks in the corresponding mass spectra (Figures S12 and S13). The m/z difference in the shifted peaks in the two sets of spectra represents the numbers of

12

C atoms replaced with

13

C atoms (cf.

Figure S14). Notably, the spectra obtained from the specimen chamber still reveal trace quantities of 12C-metabolites, what is attributed to the contaminants (ethanol peak at the m/z 47), the

12

C metabolites present in the inoculum cells, and the

12

C atoms transferred to

metabolic pathways from the 12C-amino acids and nucleobases that are present in both growth media. In addition to that, the temporal traces of

13

C-metabolites closely resemble the

temporal traces of 12C-metabolites (Figure 3B; m/z 73 and 77). It should be noted that APCI source is normally utilized for analysis of liquid samples; however, in this and some previous studies (e.g. ref.22), it has successfully been used for analysis of gas-phase samples. Previously, other MS techniques (e.g. secondary electrospray ionization4 and proton transfer reaction (PTR)21,23,24) were used to monitor VOCs emitted by microbial cultures to the gas phase. The advantage of using different ion sources is to achieve required selectivities matching the target analyte group. For example, electrospray-derived sources are typically used to ionize polar species. While the dual-chamber sampling system is

ACS Paragon Plus Environment

10

Page 11 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

not bound to any particular detection scheme, in future, it can be combined with other MS ion sources and even other detectors to attain different selectivities.

Monitoring VOCs during fructification of higher fungi Fructification of higher fungi (macromycetes) remains one of the least predictable ecophysiological processes occurring in nature. Nonetheless, understanding the mechanisms underpinning fungal fructification is of great interest to mushroom cultivators and pickers. In some saprophytic fungi, the formation of primordia and fruitbodies can be induced by specific VOCs. For instance, veratryl alcohol has been shown to stimulate fructification of oyster mushroom (Pleurotus ostreatus).25 In the case of button mushroom (Agaricus bisporus), removal of inhibitory volatile metabolites (2-ethyl-1-hexanol and 1-octen-3-ol) via increased ventilation enhances formation of primordia.26 Unfortunately, most previous studies on VOCs in higher fungi involved harvested fruitbodies (mushrooms).27-29 Typically, fungal biomass is homogenized prior to off-line analysis.30 However, such a procedure is destructive, and the information on the VOC emission kinetics is inevitably lost. In order to record data on mushroom-related VOCs in the time domain, we further verified the possibility to apply the automated dual chamber sampling system in a study involving higher fungi undergoing fructification. Mycelium of golden oyster mushroom (Pleurotus citrinopileatus) was first grown on sawdust substrate, and fructification was induced in the resulting spawn (as detailed in the Supporting Information). In this case, modified (larger) chambers were implemented to accommodate the larger size of the investigated specimens (mushroom height: < 8 cm; Figure 1E). Moreover, the flow of carrier gas in the specimen chamber had to be directed in such way that it passed through the pores in the spawn block to pick up the emitted VOCs (cf. Figure S1B).

ACS Paragon Plus Environment

11

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

The temporal profiles of the MS signals at the m/z 57 and 185 increased rapidly in the first few hours from spawn populated with juvenile fruitbodies in the specimen chamber (Figures 3C and S15). As the fruitbodies reached maturity and started to decay, those signals gradually decreased. Conversely, the profiles of the MS signals at the m/z 60 and m/z 78 showed ascending trends from ∼ 32 h (Figure S16; 6 prominent features). The profiles recorded for the control chamber (containing non-inoculated sawdust) were almost flat. The signal at the m/z 185—associated with fruitbody development—is not a pure ion but it is partly attributed to dodecanal (Figure S17). Interestingly, dodecanal had previously been found in the headspace vapor of shiitake mushroom (Lentinula edodes)31 and an extract of oyster mushroom.32 Two control experiments were further performed: (i) spawn vs. sterilized sawdust (Figures S18 and S19); and (ii) fully developed (mature/senile) fruitbodies vs. sterilized sawdust (Figures S20 and S21). These results indicate that the signals at the m/z 57 and 185 are associated with morphogenesis of P. citrinopileatus fruitbodies rather than mycelial growth or senescence. The current result is in line with the result of a previous study involving VOC extraction from pieces of harvested truffles (Tuber sp.), which showed differences in fruitbody and mycelial VOCs.33 In another interesting study, Ezra et al. monitored VOCs released by an endophytic fungus (Muscodor albus) by PTR-MS.24 To our knowledge, the current result is the first record of VOCs emitted by mycelium of a higher fungus undergoing fructification in real time. Further discovery-oriented work is warranted to pinpoint the detected VOCs to different stages of fungal development.

Real-time monitoring of food spoilage

ACS Paragon Plus Environment

12

Page 13 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

The automated dual chamber sampling system can also be used to monitor VOCs emanating from other organic matrices – for example, to track the spoilage of food-related products (e.g. meat, vegetables) during storage. To exemplify such an application, we performed monitoring of accelerated putrefaction of Pacific white shrimp (Litopenaeus vannamei). One specimen was placed in the specimen chamber, while the control chamber remained empty. Figure S22 shows representative mass spectra from this experiment, which reveal a number of putrefaction-related VOCs. Among the recorded signals, 12 signals were prominent (occurred in every replicate). Their temporal traces were dissimilar (Figure S23). They can be roughly classified into two groups according to the time when the maximum signal was registered: in the middle (m/z 49, 89, 91, 103, 105, 117, 131) or at the end (m/z 76, 90, 95, 141, 143) of the putrefaction process. Interestingly, the maxima of the five replicates do not overlap (Figure 3D; m/z 49 and 90). The observed temporal shifts are explained by biological variability (e.g. different age, life history, exposure to dissimilar environmental conditions) as well as different treatment after the harvest (e.g. exposure to different temperatures, unequal time of storage). Five of the recorded VOCs were putatively identified by taking into account the literature data on shrimp spoilage,34,35 database information36,37 (Yeast Metabolite Data Base, Human Metabolome Database), and comparative MS/MS analyses. The signal at the m/z 49 was putatively identified as methanethiol, m/z 89 as ethyl acetate, m/z 95 as dimethyl disulfide, m/z 103 as ethyl propionate, and m/z 117 as ethyl butyrate (Figure S24). In fact, formation of sulfur-containing volatile compounds and ethyl esters due to spoilage of tropical prawns and fish was previously reported.38,39 Enzymatic degradation of methionine and cysteine in seafood matrices yields methanethiol, which is further transformed to polysulfides such as dimethyl disulfide.40

ACS Paragon Plus Environment

13

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

CONCLUDING REMARKS

We have designed, constructed, and tested an automated dual chamber sampling system for tracking VOCs released by biological specimens. The platform incorporates open-source microcontrollers, which guide the operation of the experimental system. Therefore, most of its components (except detector) are inexpensive, and can be fabricated in house. The cost of this prototype amounted to ∼ 2200 USD (cf. Table S4). Apart from its labor-saving feature brought about by simple automation, the dual-chamber device synchronizes analysis of VOCs in specimen and control/blank compartments. Such synchronous analysis reduces the time required to perform both sample and control measurements, while keeping track of baseline drift. In future, the dual-chamber system can bring insights into the dynamics of VOCs emitted by biological organisms in response to various culture conditions and stimuli. While the limitation of the current study is the limited selectivity of the triple quadrupole mass spectrometer, it is appealing to couple the dual chamber sampling system with other information-rich detection platforms such as ion mobility spectrometry and high-resolution MS. It is also of interest to verify its compatibility with other MS ion sources, which are applicable in real-time VOC monitoring.

ACS Paragon Plus Environment

14

Page 15 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

ASSOCIATED CONTENT Supporting Information Available: - chamber design (SketchUp files); - additional experimental details; - microcontroller and Matlab scripts; - additional tables (S1-S4); - additional figures (S1-S24). This material is available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGEMENTS We acknowledge the Ministry of Science and Technology (MOST), Taiwan (grant numbers 104-2628-M-007-006-MY4 and 107-3017-F-007-002), the National Chiao Tung University, the National Tsing Hua University, the Frontier Research Center on Fundamental and Applied Sciences of Matters as well as the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project established by the Ministry of Education (MOE), Taiwan.

ACS Paragon Plus Environment

15

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 26

REFERENCES

1.

Audrain, B.; Farag, M. A.; Ryu, C.-M.; Ghigo, J.-M. Role of Bacterial Volatile Compounds in Bacterial Biology. FEMS Microbiol. Rev. 2015, 39, 222-233.

2.

Filipiak, W.; Mochalski, P.; Filipiak, A.; Ager, C.; Cumeras, R.; Davis, C. E.; Agapiou, A.; Unterkofler, K.; Troppmair, J. A Compendium of Volatile Organic Compounds (VOCs) Released by Human Cell Lines. Curr. Med. Chem. 2016, 23, 2112-2131.

3.

Baldwin, I. T. Plant Volatiles. Curr. Biol. 2010, 20, R392-R397.

4.

Tejero Rioseras, A.; Garcia Gomez, D.; Ebert, B. E.; Blank, L. M.; Ibáñez, A. J.; Sinues, P. M.-L. Comprehensive Real-Time Analysis of the Yeast Volatilome. Sci. Rep. 2017, 7, 14236.

5.

Wei, G.; Tian, P.; Zhang, F. X.; Qin, H.; Miao, H.; Chen, Q. W.; Hu, Z. Y.; Cao, L.; Wang, M. J.; Gu, X. F.; Huang, S. W.; Chen, M. S.; Wang, G. D. Integrative Analyses of Nontargeted Volatile Profiling and Transcriptome Data Provide Molecular Insight into VOC Diversity in Cucumber Plants (Cucumis sativus). Plant Physiol. 2016, 172, 603-618.

6.

Schnabel, R.; Fijten, R.; Smolinska, A.; Dallinga, J.; Boumans, M. L.; Stobberingh, E.; Boots, A.; Roekaerts, P.; Bergmans, D.; van Schooten, F. J. Analysis of Volatile Organic Compounds in Exhaled Breath to Diagnose Ventilator-Associated Pneumonia. Sci. Rep. 2015, 5, 17179.

7.

Bos, L. D. J.; Sterk, P. J.; Schultz, M. J. Volatile Metabolites of Pathogens: A Systematic Review. PLoS Pathog. 2013, 9, e1003311.

8.

Borjesson, T.; Stollman, U.; Schnurer, J. Volatile Metabolites and Other Indicators of Penicillium aurantiogriseum Growth on Different Substrates. Appl. Environ. Microbiol. 1990, 56, 3705-3710.

ACS Paragon Plus Environment

16

Page 17 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

9.

Morath, S. U.; Hung, R.; Bennett, J. W. Fungal Volatile Organic Compounds: A Review with Emphasis on Their Biotechnological Potential. Fungal Biol. Rev. 2012, 26, 73-83.

10. Paczkowski, S.; Schütz, S. Post-Mortem Volatiles of Vertebrate Tissue. Appl. Microbiol. Biotechnol. 2011, 91, 917-935. 11. Stalikas, C. D. Extraction, Separation, and Detection Methods for Phenolic Acids and Flavonoids. J. Sep. Sci. 2007, 30, 3268-3295. 12. Blount, B. C.; McElprang, D. O.; Chambers, D. M.; Waterhouse, M. G.; Squibb, K. S.; LaKind, J. S. Methodology for Collecting, Storing, and Analyzing Human Milk for Volatile Organic Compounds. J. Environ. Monit. 2010, 12, 1265-1273. 13. Urban P.L., Chen Y.-C., Wang Y.-S. Time-Resolved Mass Spectrometry: From Concept to Applications. 2016, Wiley, Chichester. 14. Kandiah, M.; Urban, P. L. Advances in Ultrasensitive Mass Spectrometry of Organic Molecules. Chem. Soc. Rev. 2013, 42, 5299-5322. 15. Collado, C. B.; Gomez, D. G.; Zenobi, R.; Miguel, G. V. D.; Ibáñez, A. J.; Sinues, P. M.L. Capturing in Vivo Plant Metabolism by Real-Time Analysis of Low to High Molecular Weight Volatiles. Anal. Chem. 2016, 88, 2406-2412. 16. Urban, P. L. Universal Electronics for Miniature and Automated Chemical Assays. Analyst 2015, 140, 963-975. 17. Urban, P. L. Prototyping Instruments for the Chemical Laboratory Using Inexpensive Electronic Modules. Angew. Chem. Int. Ed. 2018, 57, 11074-11077. 18. doi:10.1101/pdb.rec8247, Cold Spring Harb. Protoc. 2006. 19. Chuang, L. F.; Collins, E. B. Biosynthesis of Diacetyl in Bacteria and Yeast. J. Bacteriol. 1968, 95, 2083-2089.

ACS Paragon Plus Environment

17

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 26

20. Nordström, K. Formation of Ethyl Acetate in Fermentation with Brewer’s Yeast: IV. Metabolism of Acetyl‐Coenzyme A. J. Inst. Brew. 1963, 69, 142-153. 21. Khomenko, I.; Stefanini, I.; Cappellin, L.; Cappelletti, V.; Franceschi, P.; Cavalieri, D.; Märk, T.D.; Biasioli, F. Non-Invasive Real Time Monitoring of Yeast Volatilome by PTR-ToF-MS. Metabolomics 2017, 13, 118. 22. Chang, C. H.; Urban, P. L. Fizzy Extraction of Volatile and Semivolatile Compounds into the Gas Phase. Anal. Chem. 2016, 88, 8735-8740. 23. Bunge, M.; Araghipour, N.; Mikoviny, T.; Dunkl, J.; Schnitzhofer, R.; Hansel, A.; Schinner, F.; Wisthaler, A.; Margesin, R. On-Line Monitoring of Microbial Volatile Metabolites by Proton Transfer Reaction-Mass Spectrometry. Appl. Environ. Microbiol. 2008, 74, 2179-2186. 24. Ezra, D.; Jasper, J.; Rogers, T.; Knighton, B.; Grimsrud, E.; Strobel, G. Proton Transfer Reaction-Mass Spectrometry as a Technique to Measure Volatile Emissions of Muscodor albus. Plant Sci. 2004, 166, 1471-1477. 25. Suguimoto, H.; Barbosa, A. M.; Dekker, R. F. H.; Castro-Gomez, R. J. H. Veratryl Alcohol Stimulates Fruiting Body Formation in the Oyster Mushroom, Pleurotus ostreatus. FEMS Microbiol. Lett. 2001, 194, 235-238. 26. Noble, R.; Hobbs, P.; Pederby, J. Volatile C8 Compounds and Pseudomonads Influence Primordium Formation of Agaricus bisporus. Mycologia 2009, 101, 583-591. 27. Misharina, T. A.; Mukhutdinova S. M.; Zharikova, G. G.; Terenina, M. B.; Krikunova, N. I.; Medvedeva, I. B. The Composition of Volatile Components of Dry Cepe and Oyster Mushroom. Appl. Biochem. Microbiol. 2009, 45, 544-549.

ACS Paragon Plus Environment

18

Page 19 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

28. Misharina, T. A.; Mukhutdinova S. M.; Zharikova, G. G.; Terenina, M. B.; Krikunova, N. I. The Composition of Volatile Components of Cepe (Boletus edulis) and Oyster Mushrooms (Pleurotus ostreatus). Appl. Biochem. Microbiol. 2009, 45, 187-193. 29. Aprea, E.; Romano, A.; Betta, E.; Biasioli, F.; Cappellin, L.; Fanti, M.; Gasperi, F. Volatile Compound Changes During Shelf Life of Dried Boletus edulis: Comparison Between SPME-GC-MS and PTR-ToF-MS Analysis. J. Mass Spectrom. 2015, 50, 56-64. 30. Costa, R.; Tedone, L.; Grazia, S. D.; Dugo, P.; Mondello, L. Multiple Headspace-SolidPhase Microextraction: An Application to Quantification of Mushroom Volatiles. Anal. Chim. Acta 2013, 770, 1-6. 31. Politowicz, J.; Lech, K.; Lipan, L.; Figiel, A.; Carbonell-Barrachina, Á. A. Volatile Composition and Sensory Profile of Shiitake Mushrooms as Affected by Drying Method. J. Sci. Food Agric. 2018, 98, 1511-1521. 32. Çağlarırmak, N. The Nutrients of Exotic Mushrooms (Lentinula edodes and Pleurotus Species) and an Estimated Approach to the Volatile Compounds. Food Chem. 2007, 105, 1188-1194. 33. Splivallo, R.; Bossi, S.; Maffei, M.; Bonfante, P. Discrimination of Truffle Fruiting Body Versus Mycelial Aromas by Stir Bar Sorptive Extraction. Phytochemistry 2007, 68, 25842598. 34. Soncin, S.; Chiesa, L. M.; Panseri, S.; Biondib, P.; Cantoni, C. Determination of Volatile Compounds of Precooked Prawn (Penaeus vannamei) and Cultured Gilthead Sea Bream (Sparus aurata) Stored in Ice as Possible Spoilage Markers Using Solid Phase Microextraction and Gas Chromatography/Mass Spectrometry. J. Sci. Food Agric. 2009, 89, 436-442.

ACS Paragon Plus Environment

19

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 26

35. Giogios, I.; Kalogeropoulos, N.; Grigorakis, K. Volatile Compounds of Some Popular Mediterranean Seafood Species. Medit. Mar. Sci. 2013, 14, 343-352. 36. Yeast Metabolome Database, http://www.ymdb.ca/ (accessed on 3rd August, 2018). 37. Human Metabolome Database, http://www.hmdb.ca/ (accessed on 3rd August, 2018). 38. Chinivasagam, H. N.; Bremner, H. A.; Wood, A. F.; Nottingham, S. M. Volatile Components Associated with Bacterial Spoilage of Tropical Prawns. Int. J. Food Microbiol. 1998, 42, 45-55. 39. Duflos, G.; Coin, V. M.; Cornu, M.; Antinelli, J. F.; Malle, P. Determination of Volatile Compounds to Characterize Fish Spoilage Using Headspace/Mass Spectrometry and Solid‐Phase Microextraction/Gas Chromatography/Mass Spectrometry. J. Sci. Food Agric. 2006, 86, 600-611. 40. Varlet, V.; Fernandez, X. Sulfur-Containing Volatile Compounds in Seafood: Occurrence, Odorant Properties and Mechanisms of Formation. Food Sci. Technol. Int. 2010, 16, 463503.

ACS Paragon Plus Environment

20

Page 21 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

FIGURE CAPTIONS

Figure 1. Schematic representation of the automated dual chamber sampling system for monitoring VOCs emitted by biological specimens (A); photograph of the assembled system (without incubator and detector; B); photographs of the studied specimens, L. vannamei (C); S. cerevisiae (D); and P. citrinopileatus (E).

Figure 2. Typical raw MS data subjected to processing. A Petri dish with S. cerevisiae (23-h after inoculation) was inserted to the chamber, and data were recorded: (A) total ion current; (B) extracted ion currents (blue line, m/z 87; red line, m/z 89); (C) control chamber spectrum; (D) specimen chamber spectrum. In (C) and (D), red bars highlight the specimen-related features, while blue bars highlight the control (blank)-related features (no baseline subtraction).

Figure 3. Representative results obtained with the automated dual chamber sampling system operated with positive-ion APCI-QQQ-MS: (A) S. cerevisiae cultured on

12

C-glucose

agar medium (monitored simultaneously with blank – uninoculated medium; n = 3); (B) S. cerevisiae cultured simultaneously on (blue line –

12

C chamber; red line –

13

12

C-glucose and

13

C-glucose agar media

C chamber; one out of three replicates);

(C) P. citrinopileatus spawn in the process of fructification on supplemented sawdust (n = 3); (D) L. vannamei specimen undergoing putrefaction (n = 5).

ACS Paragon Plus Environment

21

Analytical Chemistry

Table 1. Steps in the analysis of VOCs emitted by biological specimens using the automated dual chamber sampling system with nitrogen or air as carrier gases. Step

Chamber Step

Action

Recording detector baseline

2′

Injecting control chamber headspace vapors

3′

Control / blank

1′

Flushing tubing with fresh air

4′

Flushing tubing with fresh air via control chamber

1′′

Recording detector baseline

2′′

Injecting specimen headspace vapors

3′′

4′′

Specimen

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 26

Flushing tubing with fresh air Flushing tubing with fresh air via control chamber

- no carrier gas flow - detector is isolated from the chamber - carrier gas flows to the chamber - carrier gas moves headspace vapors to the detector - switching off data acquisition - switching on air pump to remove residues in tubing - replenishing air required for the growth of microorganisms - removing ethanol residue from tubing* - no carrier gas flow - detector is isolated from the chamber - carrier gas flows to the chamber - carrier gas moves headspace vapors to the detector - switching off data acquisition - switching on air pump to remove residues in tubing - replenishing air required for the growth of microorganisms - removing ethanol residue from tubing*

yeast

Duration / s mushroom

shrimp

15

20

15

75

280

45

120

120

120

60

0

0

15

20

15

75

280

45

120

120

120

60

0

0

- closing all pinch valves to accumulate volatile compounds 3870 5400 3780 released by microorganisms * The amount of ethanol produced during the growth of yeast cells is high, and it causes contamination of the sample duct. Therefore, an additional step of flushing tubing with pure air has been introduced. 5

Waiting for next run

ACS Paragon Plus Environment

22

Page 23 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 1 (one column)

ACS Paragon Plus Environment

23

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 26

Figure 2 (one column)

ACS Paragon Plus Environment

24

Page 25 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 3 (one column)

ACS Paragon Plus Environment

25

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 26

Graphical abstract

ACS Paragon Plus Environment

26