View all hidden authors and organizations

Edited by Alexis T. Bell, University of California, Berkeley, approved on August 25, 2021 (review received on April 20, 2021)

The widespread use of algae cultivation to produce environmentally sustainable biofuels and fine chemicals is currently hampered by the massive losses (10% to 30%) caused by herbivore infections. We demonstrate the use of real-time chemical ionization mass spectrometry to quickly identify gaseous indicators for herbivore infection in cyanobacteria cultures. Compared with current microscopy and qPCR methods, the detection of grazing by real-time mass spectrometry is much faster (up to 3 days). By adopting this technology, growers can deal with herbivore infestations faster, thereby protecting crops and increasing profitability.

The cultivation of algae in open waterways ponds is considered to be the most economical way to produce biomass for biofuels, chemical feedstocks and other high-value products through photosynthesis. One of the main challenges of opening ponds is the reduction in biomass production due to attacks by herbivores, competitors, and infectious organisms. More frequent observations are needed to detect herbivore infections, which can quickly reduce biomass levels. In this study, chemical ionization mass spectrometry (CIMS) was used for real-time measurement to monitor the effect of herbivore infection on cyanobacteria culture. Freshwater Synechococcus elongatus Pasteur Culture Collection (PCC) 7942 produces a lot of volatile gases during healthy growth, among which 6-methyl-5-hepten-2-one is a unique metabolic indicator of exponential growth. After the introduction of Tetrahymena ciliate, the concentration of various volatile substances was observed to change after a short incubation period of 18 hours. It has been found that nitrogen-containing gases, including ammonia and pyrroline, are reliable indicators of grazing. The detection of grazing by CIMS is much faster than traditional methods, microscopic examination, and continuous fluorescence detection. The traditional method does not detect changes until 37 to 76 hours after the CIMS detection. The CIMS analysis of the gas produced by PCC 7942 further shows the complex time array of biomass-dependent volatile gas production, which demonstrates the potential of using volatile gas analysis as a diagnosis of herbivore infection. Overall, these results show the prospect of using continuous volatile metabolite monitoring to detect single algae grazing, which has the potential to reduce the biomass loss caused by current grazing, which can save hundreds of millions of dollars.

Microalgae are prokaryotic or eukaryotic photosynthetic organisms that can grow rapidly on a large scale under various conditions. Compared with terrestrial crops such as corn or soybeans, their efficiency in producing large amounts of biomass in a small space makes microalgae a promising and sustainable platform for the biological production of fuels or industrial products. In addition to the most popular product, biodiesel, microalgae can also be used to produce other valuable fine chemicals, such as dyes, cosmetics, drugs, and food additives (1⇓ –3). Alternatively, microalgae can also be used in conjunction with other industrial processes, such as removing dissolved nutrients from wastewater and removing CO2 from flue gas or the atmosphere (4, 5). So far, algae research has focused extensively on the improvement of crop productivity, including strain selection, nutrition control, and management of physical and chemical parameters (pH, temperature, etc.). To achieve this goal, many growers use large-capacity open runway ponds (ORP) instead of the more expensive laboratory photobioreactors (PBR) (6). Although the cost of algal biomass from ORP is about an order of magnitude lower than PBR, a key disadvantage is the increased risk of contamination by unwanted herbivores, such as ciliates, rotifers, viruses, bacteria, and fungi (7, 8). Recent estimates of commercial monocultures in ORP found that crop failures caused by herbivores reduced production biomass by 10% to 30%, resulting in hundreds of millions of dollars in losses (9). Part of the challenge posed by herbivore contamination relates to their ability to eradicate microalgae cultures quickly, sometimes within 48 hours of contamination (10). An integral part of proper integrated pest management is the use of technologies that can quickly detect herbivores at the lowest possible pest concentration (11). Once growers are notified of the infection, they can choose to treat the culture to kill or slow the growth of contaminants, if there is supplemental herbivore identification information, or immediately harvest to save the crop (3).

Many analytical methods have been used to detect herbivores (3, 12, 13). Microscopy and automated optical techniques, such as flow cytometry, have made progress in detecting low-concentration (<10 units/ml) herbivores (14, 15). Although simple and cost-effective, these techniques are offline, can be slow to operate, and require specific protocols or knowledge about the appearance of herbivores. Alternatively, techniques from molecular biology, such as qPCR, have shown promise in identifying herbivores, with a theoretical detection limit of a single molecule; however, the challenges associated with cost and the production of harmful herbivore-specific primer libraries are significant Limit (12, 16). It is worth noting that these two methods are very susceptible to sampling bias. Some herbivores prefer to be located on surfaces or biofilms, so they cannot be identified because most samples are collected from liquids.

Recently, people have focused attention on the series of volatile gases emitted by microalgae during their prosperous life cycle (17⇓ –19). For algae, the emission of volatile gases is mediated by their environmental conditions or biological states, such as exponential growth, nutrient availability, photooxidative stress, and aging (17). In addition, the presence of bacteria has been shown to transform and regulate the volatile gases and metabolites produced by microalgae (20, 21). In general, the "volatiles" of microalgae represent a reflection of the state of health (SoH) of the organism, and because of the significant impact of grazing on the health of microalgae, it reflects the infection status of water blooms. Therefore, the development of techniques and instruments to measure microalgae volatiles represents an opportunity to identify herbivore infections in a highly sensitive and descriptive manner (17). In addition, the problem of sampling deviation is less, because the distribution of volatile gases in the solution is much better than physical predators fixed on the surface of the container. At present, the main technique for analyzing algae volatiles is gas chromatography-mass spectrometry (GC/MS). Although the identification of volatile gases is extremely sensitive and effective, ordinary GC/MS systems are not suitable for continuous in-situ measurements without major modifications (22). In addition, solid-phase microextraction (SPME), the main sampling technique for microalgae volatile gases, usually requires a long-term equilibration of up to 24 hours to obtain sufficient signals (17). Unfortunately, such a long sampling time not only delays detection, but also averages transient and potentially important changes in the concentration of volatile gases. In view of the rapid time scale (1 to 2 days) of herbivore infection, headspace monitoring technology with high time resolution and the ability to easily switch and measure between multiple blood vessels is required.

Chemical ionization mass spectrometry (CIMS) is a chemically selective method commonly used in the atmosphere, food, medicine, national defense, and drug law enforcement disciplines for online detection of various gas-phase species with heterogeneous spatiotemporal concentrations (23). In CIMS, the ionization reagent gas is mixed with a continuous sampling gas stream from the sample headspace. The favorable chemical energy between reagent ions and analyte molecules in the sample headspace causes soft ionization of the analyte, which is then detected by various types of mass spectrometry. It is worth noting that this method can bypass the need for gas chromatography columns, because the selection of specific reactive ions (such as H3O+, C6H6+ or I-) can be used to select specific types of molecules while eliminating interfering air from the main components. (N2, O2, CO2, Ar, CH4, etc.) are usually ionized by electron impact in GC/MS (24). The removal separation step can measure ambient gas in a highly continuous (1 Hz or faster) manner, and the soft nature of chemical ionization limits ion fragmentation, which makes the mass spectrum of unseparated gas too complicated. The use of the most common form of chemical ionization, proton transfer reaction mass spectrometry (PTR-MS), studies of volatile gas emissions in natural marine and freshwater systems have shown that CIMS can identify different biological activity states of phytoplankton and bacteria (25, 26). These findings indicate that CIMS may have the ability to monitor highly concentrated single cultures of algae and benefit from far higher throughput than modern GC/MS technology. In fact, the use of CIMS and PTR-MS has been applied to the continuous monitoring of fermentation biological processes and the crop herbivory detection of harmful parasites in dozens of other use cases (27⇓ ⇓ –30).

In this study, we used CIMS to monitor the monoculture of freshwater cyanobacteria before and after adding the field-isolated tetrahymena ciliates PCC 7942. In the 28-day experiment, CIMS continuously monitored three 20-liter rats through corresponding biological measurements because they continued to be infected by Tetrahymena during their healthy growth. Switching between sampling containers at 15-minute intervals allows the generation of algae volatile gases to be detected with a higher time resolution than previously used, allowing direct comparison of the ability of gas analysis to detect herbivore infections compared to traditional techniques. The analysis of CIMS data reveals unique insights into the time scale of gas production, highlighting the potential for high time resolution gas analysis of SoH and herbivore infections in microalgae cultures.

In order to prove the ability of CIMS to analyze cyanobacteria SoH and grazer infections, we simultaneously monitored three 20-L carboy cultures of S. elongatus PCC 7942 (hereinafter referred to as PCC 7942) through sterile exponential growth and subsequent planned field infections -The isolated Tetrahymena quickly prey on PCC 7942. For comparison, a liquid sample is collected from each tank at least once a day for microscopic examination, absorption spectroscopy, and fluorescence spectroscopy. After adding the predator, a continuous fluorescence spectroscopy system was also used to monitor infected great white sharks. The culture is continuously bubbled with sterile zero air, pushing the headspace gas above the culture to a customized solenoid valve array, which is programmed to be between each of the three large bottles and direct zero air input every 15 minutes Switching (Figure 1B). The headspace or zero air sample source activated by the solenoid valve array is pulled directly into the ion molecule region (IMR) of the chemical ionization time-of-flight mass spectrometer, where the continuous air flow from the large bottle or zero air is removed by (H2O )nH+ ionization is carried out by proton transfer and directly introduced into CIMS without any column chromatography (Figure 1A). The mass spectrum is accumulated as the sum of 60,000 spectra collected in 1 second (Figure 1C).

The experimental arrangement of the Carboy infection experiment. (A) CIMS instrument diagram, (B) Carboy sampling setup with solenoid valve array, (C) Example sampling plan of solenoid valve array for switching between Carboys for CIMS headspace sampling.

The mass spectrum collected from the headspace on the PCC 7942 culture showed many peaks indicating the presence of volatile gases in the sample. Figure 2A shows a comparison of the average mass spectra obtained from Carboy 1 headspace and clean zero air during the growth of sterile algae on the fourth day after PCC 7942 inoculation. The mass spectra of both sample types are dominated by water cluster reagent ions with mass-to-charge ratios (m/z) 19, 37, 55, 73, and 91, which are typical clusters formed in chemical ionization. In all large bottles, the rest of the mass spectrum consists of numerous ions produced by proton transfer (M+1), which fall between m/z 40 and 200. At this time, the ionic strength fraction of non-aqueous cluster ions occupied by zero air is 7%, and the headspace of Carboy 1 is 11%, and the total ion count (TIC) is 2.5 × 106 counts per second. The ionic strength ratio of non-aqueous cluster ions between Carboy 1 and zero air is 1.51. In this ionic strength, Carboy 1 has 75 unique m/z, with an average strength 25% higher than zero air, indicating that they are from cyanobacteria production, BG-11 growth medium or Carboy components. Among these 75 unique m/z, 49 ions were found to exhibit obvious time-varying behavior during the experimental life of Carboy 1. This time-varying behavior is operationally defined as an increase or decrease in the normalized intensity over 10% The time scale of the experiment. The time series showing exponential decay from the beginning of the experiment or spikes caused by pressure instability were excluded from this analysis, indicating that the 49 ions that changed intensities during the life of the culture were most likely due to active organisms. Process rather than shut down-release gas from the medium or experimental device. These results together prove that the water cluster CIMS can detect a wide range of unique volatile gases, which may be related to the health of cyanobacteria and other microalgae cultures.

(A) Average CIMS mass spectrum of zero air and Carboy 1 headspace. (B) 1/e2 histogram of ion equilibrium time obtained from switching between sample zero air and Carboy 1.

Among the ions detected by CIMS, complementary MS methods such as solid phase microextraction gas chromatography mass spectrometry (SPME-GC/MS) and modified atmospheric pressure chemical ionization high resolution mass spectrometry (APCI-Orbitrap) are combined for direct analysis The latest innovation of gas phase species (31) (Protocol S1). These identifiers are listed in Table S1 of the SI Appendix. In general, the identification of the molecular formula or hypothesis is performed on a small fraction (<15%) of the total number of ions observed by CIMS during the experiment. It has been observed that many of these species are aliphatic ketones and aldehydes, similar to those observed by others in algae surveys (17, 18, 32⇓ –34); however, nitrogen-containing gases including ammonia and C4H7N are top Although there is evidence that phytoplankton in natural seawater can produce methylated amines, this is an important part of the void, but this is usually ignored in the previous analysis of volatile gases from commercial algae (33, 35, 36).

Although the CIMS used in this work has relatively low mass resolution (~1,200 full width at half maximum) and does not cause significant ion fragmentation that can be used for molecular identification, it can be valuable by monitoring the rise or decay time The analyte information of the CIMS is the change in ions when the CIMS switches between sampling containers. This rise or decay time (SI appendix, Figure S2) can be modeled using double exponential fitting, and the synthesis time (37, 38) for the ion to reach 1/e2 (86.5%) of its equilibrium intensity can be calculated. The 1/e2 decay time is reflected as the sum of multiple physical processes because the gas flows from the sample source to the IMR inlet of the instrument along the length of the pipe. Molecules with a high saturation concentration (volatility) remain in the gas phase, which results in a shorter equilibration time between samples. Conversely, low-volatility molecules tend to stick to the pipeline and cause longer equilibration times (37). In addition to the saturation concentration, smaller molecules with high dipole moments, such as NH3 or HCl, may be distributed into the water microlayer on the tube wall, thereby significantly increasing their equilibrium time (38). In summary, the analysis of the gas rise or decay time, combined with their corresponding m/z, provides valuable information about the general chemical properties of the observed ions. This analysis can also calculate the minimum time that CIMS must monitor the sampling vessel before switching to the next sampling vessel.

When CIMS switches between vial and zero air, we use the rise or decay time to apply this analysis to a set of 42 unique m/z. The analysis results of the data collected on the fourth day after Carboy 1 was inoculated with PCC 7942 are shown in Figure 2B, with only the 1/e2 value of ammonia (NH3) omitted. It is worth noting that most of the 1/e2 decay time is less than 12 seconds, which is equivalent to the equilibration time required for high-speed environmental measurement (39). In contrast, ammonia and other even-numbered mass gases with m/z of 70 and 76 have even-numbered m/z values ​​indicating that if ionized by proton transfer [(H2O)nH+ the main ionization scheme], there will be an odd number of nitrogen Atom, showing longer equilibration time (14 and 19 seconds), ammonia 1/e2 time is about 335 seconds. The distribution of nitrogen in the molecular formulas of these species is consistent with their longer equilibration time, because their significant alkalinity promotes dissolution in the water on the pipe wall. Based on the maximum ammonia equilibration time of 5.6 minutes, this key operating parameter means that CIMS and a proper sample switching device can analyze 10 different containers in one hour. In addition, these equilibration times are not fixed and can be further improved by higher flow rates, pipe diameters, heating, and chemical coating of the pipe or CIMS inlet. Finally, the analysis showed that the 15-minute sampling plan used in this experiment was sufficient to accurately determine the ionic strength of the various gases present in the culture headspace sample.

Volatile gases from Carboys 1, 2 and 3 were monitored for 28 days. All Chinese cabbage was inoculated with PCC 7942 at the same time, and each Chinese cabbage was infected with Tetrahymena at a concentration of 0.1 cells/ml on the 8, 13 and 23 days (Figure 3, vertical solid line). This infection density was chosen because it is well below the typical limit of detection (LOD) for microscopy (5 to 20 cells/ml) and the theoretical blank limit of the technique used (20 cells/ml (40⇓ –42)). Twice a day after infection, manually collect liquid samples from the infected large bottles and analyze the presence of Tetrahymena under a microscope. Herbivores were first observed on days 12, 16, and 27 (Figure 3, vertical dashed line), and more information about the specific duration before testing is summarized in Table 1. Although qPCR was not used to detect herbivores in these experiments, a separate analysis was performed to evaluate the LOD for qPCR detection of Tetrahymena. For details, see the SI appendix. The observed qPCR LOD of 597.6 cells·mL-1 of Tetrahymena (SI Appendix, Figure S4) greatly exceeded the detection threshold of the microscope; therefore, we chose the microscope as the benchmark herbivore detection method during the experiment time. In order to observe the changes in algal biomass associated with grazing on a faster time scale, each inoculated herbivore was monitored by continuous fluorescence (Figure 3B) until the biomass was completely lost, such as daily fluorescence measurement, visible color The changes and culture collapse are shown (Figure 3B). 3A). Although continuous fluorescence provides higher time resolution in displaying biomass loss during collisions caused by herbivores, herbivore detection does not have any significant advantages compared to manual daily fluorescence sampling. In addition, during the infection of carboy 3, the continuous fluorescence signal began to gradually weaken before the manual daily sampling changed, which may be due to the observation of biofouling of the sensor under high algal biomass. The optical density (OD) at a wavelength of 750 nm is also monitored daily (SI appendix, Figure S5). Although OD and fluorescence showed good agreement in these experiments, fluorescence was chosen as the main measurement because it better reflects the presence of living cyanobacteria, rather than OD, which can capture senescent cells and debris. Given the consistency between OD and fluorescence, we found that in these experiments, any self-masking during growth may not completely change the production of chlorophyll, which may distort the evaluation of culture productivity.

Carboy colored time series data of the Carboy experiment (Carboy 1: green, Carboy 2: red, Carboy 3: blue). (A) Daily fluorescence: 590 nm excitation/670 nm emission. Arbitrary unit, AU. (B) Continuous flow cuvette fluorescence: 420 nm excitation/670 nm emission. (C) CIMS m/z 18 (NH3) intensity normalized counts per second, ncps. (D) CIMS m/z 137 (monoterpene) intensity. (E) CIMS m/z 70 (C4H7N) intensity. The vertical solid line represents the addition time of Tetrahymena per barrel; the vertical dotted line represents the first detection of Tetrahymena through a microscope.

Time after adding herbivores before the detection technology marks the presence of Tetrahymena

Many gases show significant changes in intensity in response to herbivore infections; however, the time series of the species with the fastest response and the highest intensity will be discussed here. The most significant change in ionic strength is NH3 at m/z 18 (Figure 3C) and its related water cluster adducts at m/z 36 and 54. It was observed that the initial NH3 ionic strength increased at the beginning of the experiment, and it was an indication of the volatile gas of NH3 containing 6 mg/L ferric ammonium citrate in the BG-11 medium observed by monitoring the sterile BG-11 (SI appendix, Figure S6). After the first 2 days of algae growth, the NH3 ion signal decreased by nearly two orders of magnitude, and remained below 2 × 104 counts s−1 until after Tetrahymena was added. This drop is most likely due to the absorption of ammonium by PCC 7942 and the gradual ventilation of NH3 through bubble zero air. After adding a grazer to each Carboy, the initial NH3 intensity peak of each Carboy was observed on the 11th, 15th, and 25th days. It is worth noting that during Carboy 1's ammonia peak, some NH3 signals were transferred from Carboy 1 to Carboy 2 and Carboy 2 was transferred from Carboy 2 to Carboy 3 during Carboy 2's ammonia peak. These residual events were caused by excessive ammonia in the sampling line and CIMS inlet, which did not evaporate sufficiently before the CIMS switched to the next sampling vessel, which was consistent with the observed slow ammonia equilibrium rate. After the first intensity spike, the NH3 signal decreased before the second spike, which was consistent with culture collapse and decreased fluorescence signal. It is worth noting that as the duration of the experiment continued, the intensity of the first ammonia peak increased, which was consistent with the increase in algae biomass in the later period (Figure 3A and C). We tentatively suggest that the size of the first ammonia peak seems to depend on biomass and may be a response related to algal stress. Although ammonia release from certain Anabaena strains has been described (43), although not as a stress or predation response, algae growers have used the addition of free ammonia as a method of controlling pests (44). We hypothesize that this initial ammonia peak represents a possible defense mechanism. Contrary to the first peak, the magnitude of the second ammonia peak is relatively consistent for all three carboys, indicating that its production is a function of biomass degradation caused by herbivores' predation of cyanobacteria.

Although the ammonia signal increased after the addition of predators, the intensity of m/z 137 (an unidentified monoterpene) showed a significant decrease in Carboys 1 and 2 several days before the detection of Tetrahymena through the microscope (Figure 3D). For Carboy 3, the strong NH3 signal from the first spike disrupted the ion chemistry of CIMS and prevented reliable observation of m/z 137 (SI appendix, Figure S7). Therefore, if the m/z 18 intensity exceeds 2,500 counts per second, the m/z 137 intensity will be discarded. In addition to nitrogen-containing (uniform mass) ions, most other ions also follow this protocol, and these ions show intensity stability during periods of high NH3. This stability may be due to the high gas phase proton affinity of these molecules. If NH3 has been fully titrated (H2O)nH+ as an active reactive ion, they can receive protons from NH4+ (proton affinity [PA] = 853.6 kJ/mol); However, monoterpenes (PAalpha pinene = 878 kJ/mol) will not. In Figure 3E, the intensity of m/z 70 is determined to have the molecular formula C4H7N and may be 1- or 3-pyrroline (PA = 925.8 or 931.0 kJ/mol). For infected Carboys, the intensity shows a strong positive Changes 2 and 3 but only reduced intensity in Carboy 1 (Figure 3E, inset). Similar to the observed double spikes of NH3, the algal biomass present at the time of infection seems to be an important factor in the response to grazing to produce m/z 70. These results are further supported by the measurement results of m/z 32 and methylamine (PA = 899.0 kJ/mol), which only appeared in the grazing response in Carboy 3, which reached before the collision caused by the herbivore. The highest biomass level (SI appendix, Figure S8). In view of the destruction of (H2O)nH+ ion chemistry by NH3, in these experiments, the authors pointed out that sample dilution is an option for future analysis to counteract the titration of NH3; however, this is the sensitivity to non-NH3 analytes in the sample stream These analytes are also diluted.

Observing the characteristics of volatile gases that continue to change with the growth of all cultures after the introduction of predators, we compared the time of gas intensity changes detected by CIMS to show the time relative to microscopy and continuous fluorescence detection of herbivores , The presence of herbivores. By applying the detection threshold based on ion signal changes of 10σ within a 4-hour window, we found that CIMS detects the effects of herbivores 25 to 76 hours before the microscope inspection, depending on the Detected gases and specific experiments (Table 1). In all cases, C4H7N is the first gas to appear or decrease due to grazing and is effective for all three carbons, while monoterpenes are more effective at lower biomass. In this case, high ammonia causes There are fewer complications. It is worth noting that the duration of herbivore detection achieved by volatile gas analysis is reduced during the duration of the experiment, indicating that the increase in biomass in Carboys 2 and 3 seems to shorten the algae's response to infection. For microscopy, the detection of herbivores after Carboys 1 and 3 were very similar; however, Carboy 2 happened earlier. The high variance of the microscope inspection time reflects the inherent challenges of this technique at low cell densities, in which the statistical fluctuation and heterogeneous localization of predators may be prominent factors. For example, a 1ml sample that screens herbivores with a detection limit of 1 cell/ml may not contain herbivores, even though the concentration of bulk herbivores is about 1 cell/ml. Although continuous fluorescence cannot detect herbivores until the cyanobacteria biomass is reduced, the very similar time reflects the herbivorous consistency between the three Carboy inoculations, which helps to verify the chemical comparison between the three systems.

In order to better understand whether the obvious correlation between the change in the volatile gas intensity of certain chemical substances and the increase in cyanobacteria biomass is a more common phenomenon, we calculated the 1-h derivative of the movement of the ion signal, which varies with grazing Changes and evaluates these signals to be changed by a separate carboy. From the analysis of the maximum raw ionic strength derivative of the gas in each bottle (Figure 4A), it is obvious that a denser algae culture will produce a larger total gas change in response to grazing, which indicates that the use of algae gas as a diagnosis will It is more effective for denser, mature ponds. In addition, the maximum derivative of each ion is normalized to the respective original intensity when Tetrahymena is added, and then the time for the ion to double in intensity at this rate of change is calculated (Figure 4B).

Derivative analysis of gas production affected by grazer from PCC 7942. (A) The maximum derivative value of the original ion signal change. (B) The normalized derivative doubling time of the same group of ions in A. The × mark indicates the average signal derivative and doubling time of each large bottle. Compared with other ions observed in this study, the ions selected from Figure 3 are represented by colored boxes in each panel to emphasize their time-dependent behavior.

This calculation allows to investigate how much each volatile gas has changed over time as a response to the addition of the grazer between the three large bottles. The results in Figure 4B show that the gas doubling time is significantly reduced from Carboy 1 to Carboy 3, indicating that the initial amount of algal biomass at the time of infection accelerates the production or loss of gas in response to grazing. In addition, this observation indicates that as the doubling time decreases, a higher time resolution analysis method, such as CIMS, is required to capture these signal changes in the manner of current GC/MS-based gas analysis technology for effective herbivory. Animal testing cannot be done.

Although the main scope of this study focused on the changes in gas production in algae infected by herbivores, we also observed interesting changes in the volatile gas intensity of PCC 7942 during healthy growth. As shown in Figure S9A in the SI appendix, a set of demonstration gases plots the time series of the entire experiment in Carboy 1. Each of these gases increases in intensity with the growth of PCC 7942 in Carboy 1; however, the different shape changes in the intensity of each ion during the growth process prove the biochemical complexity that may be the basis for each compound. It is worth noting that m/z 127, 6-methyl-5-hepten-2-one (6-MHO) showed a continuous increase in algae growth, reaching about 5.5 days in the experiment and maintaining roughly the same signal intensity , Until the grazing pressure of Tetrahymena (the 8th day of addition) on PCC 7942 resulted in the loss of productive algal biomass capable of producing 6-MHO. Due to continuous bubbling throughout the experiment, resulting in continuous ventilation of dissolved gases, the 6-MHO platform indicated that the algae production of 6-MHO reached a steady-state value that may have biological information. In fact, the deviation of 6-MHO signal growth seems to correspond to the change of algae culture productivity, in which the exponential algae growth almost stabilized during the experiment time (SI appendix, Figure S9B). In order to further compare the volatility in Carboy 1 For gas production and healthy growth, the daily average ionic strength of the chemical species shown in Figure S9A of the SI Appendix is ​​divided by the daily average algae fluorescence. As shown in the SI appendix, Figure S9 A, B, DF, the ratio of each ion to daily fluorescence drops to a stable value until the algal biomass is degraded by grazing after the 8th day. The stability of the ion to fluorescence ratio indicates that the productivity of these gases per cell may remain constant. However, the ratio of 6-MHO to fluorescence exhibited a different behavior throughout the experimental time, which was also consistent with the deviation from exponential growth to decay growth (SI Appendix, Figure S9B). Although it is beyond the scope of this work to evaluate the specific biological purpose or cause of the release of each volatile gas, the ratio of ionic strength to algae fluorescence can help distinguish whether the per-cell yield of volatile substances increases with biomass. And change. We believe that better distinguishing the effects of biomass variability and invariant volatile gases may open many new doors for a better understanding of algal biochemistry in the future. Although it is necessary to further study the relationship between PCC 7942 growth and 6-MHO production, these results indicate that some volatile gases can be used as a metabolic indicator during the growth phase of the culture to better inform growers how to manage algae crops to maintain Or change the growth period to maximize biomass productivity or induce the production of desired products.

Several promising molecular species have been identified using chemical ionization mass spectrometry as diagnostic tracers to identify Tetrahymena grazing in PCC 7942. We found that nitrogen-containing gas is a highly reliable and unique tracer, indicating algae grazing. Although it is not possible to make a hypothetical identification of all ions, additional analysis using GC/MS and atmospheric pressure chemical ionization-high resolution mass spectrometry (APCI-HRMS) is used to obtain more information. However, in order to identify predation or infection in algae cultures, it is not necessary to determine the identity of all volatile gases measured by CIMS. If the m/z measured by CIMS always responds to destructive pollution, knowing its identity and biochemical purpose will only give a better understanding of the biochemistry of the entire system. This is similar to medical measurements of temperature or blood pressure, which are related to pathogenesis in a generally unknown and highly complex system and can be used to indicate changes in the patient's health.

In addition, this study evaluates and compares volatile gas measurements with standard monitoring techniques (such as microscopy and spectroscopy) based on time-based key indicators. These indicators include 1) the detection time after LOD infection, which reflects the method’s The number or level of grazing activities with sensitive pollutants, or 2) the detection time before the farming collapse, is an important criterion for growers, as it determines how much time can be used to prevent or restore potential biomass loss. The data provided in Figure 3 and Table 1 prove that the performance of CIMS real-time measurement is better than that of microscopy and spectroscopic monitoring methods. The analysis of volatile gases in the PCC 7942 culture infected by herbivores provides at least 24 hours of detection characteristics. For this article The experimental conditions discussed are faster than microscopes, up to 3 days. Although the custom CIMS type and (H2O)nH+ chemicals used in this work can effectively detect herbivore infections, other types of chemical ionization instruments, such as proton transfer reaction mass spectrometry (PTR-MS) and selective ion flow tube mass spectrometry ( SIFT-MS), should be equally useful. Both PTR-MS and SIFT-MS are available on the market, and although they are expensive, they have enhanced capabilities that enable wider selectivity and reduce humidity dependence.

We further determined the fascinating characteristics of the time scale and intensity of volatile gases in response to grazing, which appeared to be related to algal biomass. First, we observed short-lived early signal changes, the intensity of which changed disproportionately as the biomass in each tank increased. For example, the peak maximum intensity of the first ammonia burst in Carboys 1 and 2 differed by about 25 times (Figure 3C, days 11.5 and 15.5), while the fluorescence or OD changes were both 1.1 times. Based on the transient and disproportionate biomass-dependent nature of this response, we hypothesized that this feature is unlikely to be directly caused by biomass degradation or increased encounter rate, but may represent a defensive response that is Non-linear amplification of denser cell populations. The possibility of algae to respond to stress stimuli in a population-dependent manner, as observed in the literature of abiotic stress (45), suggests that gas analysis as a diagnostic method may require multiple data types to contextualize The overall SoH algae crop. Second, we observed that when the infection occurred at a higher initial biomass density, the overall signal change occurred earlier (Figure 4). Since this phenomenon is more commonly observed in all researched volatile signals (including non-transient signals), the cause is more likely to be a direct consequence of biomass degradation and is related to increased encounter rates in high-density cultures. Subsequent predator replication increases and grazing rate. Interestingly, the microscope and fluorescence detection time have not changed in a similar way. The ability of ponds to respond to stress stimuli in a population-dependent manner tells us that gas analysis as a diagnostic method may require multiple data types to contextualize the overall SoH of algae crops. We also used our high time resolution measurement results to show that the gas composition changes caused by grazing occurred on a <24 hour basis, and in less than an hour, some intensity changes were 2 times or more. The sampling frequency of algae volatile gas diagnostic measurement may need to occur more than once a day, ideally once or more per hour, which is not easy for GC/MS (a typical method used in this type of research) . In view of the restrictions imposed by the distribution of volatile gases to the sampling tube, we found that CIMS can monitor about 10 containers per hour, which will be improved by modifying the sampling schedule for field applications, which greatly offsets the cost of this more complex instrument . The future application of this method in algae monitoring will significantly benefit from more complex numerical analysis, which can link algae gas production with various other health indicators, so that farmers can "determine" the pond's status. The ability to be infected in automation. Way. The grower can then initiate scheduled interventions in an automated manner to minimize biomass loss, such as those observed in these experiments and on-site. In view of the periodic reports of 30% or more loss of biomass production and the notification of any major biomass loss that may occur within the CIMS system 1 to 3 days in advance, we believe that the CIMS monitoring system can effectively allow growers time to recover almost All the lost biomass is harvested even with the most drastic intervention of the entire system.

Future work on this approach will require the study of multiple algae-herbivore pairs to assess the specificity of volatile gas responses to grazing. Given that many microalgae are known to produce different volatile metabolites, some chemicals may be unique to algae-herbivore pairs (17). Other gases, which may be simpler molecules such as NH3 derived from common precursors such as proteins, may be preserved in many algae-herbivore combinations. The development of future field applications will need to solve the sampling problem of unwanted outdoor pollutants, which may confuse the analysis of crop volatiles. Here, we envisage the use of a sampling interface, which provides a large surface area of ​​pond water sampled by a clean air stream, which is measured by CIMS. This method can at least protect the system from drastic changes in the intensity of the gas carried by the local air source, but it is still susceptible to gaseous compounds that may enter the pond over time. Therefore, it is necessary to carefully select diagnostic markers for herbivore infection based on the risk of external contamination of similar or isobaric compounds. Although the analysis in this study focuses on the volatile gas assessment of microalgae health, it is important to note that chemical ionization mass spectrometry can be used to monitor health indicator gases in many other production systems, including beer, cheese, monoclonal antibodies, and experiments All of the meat cultivated in the room requires strict cleaning.

S. elongatus PCC 7942 in BG-11 medium (46) in BG-11 medium (46) in a 250-mL or 2-L flask at 30°C with continuous shaking (125 rpm) at 200 μmol Grow under continuous light to maintain the photons of the tetrahymena from the fluorescent cool white bulb ⋅ m−2 ⋅ s−1.

As described by Schoepp et al., 30 liters of S. elongatus PCC 7942 BG-11 culture grown in 100 liter plastic bags at the University of California San Diego Greenhouse Biology Field Station. (47) Crash due to unknown pollutants. A sample of this broken culture was brought into the laboratory and analyzed by an optical microscope. The main organism was observed by the optical microscope as free-swimming ciliates. Dilute the culture sample serially or manually separate it under a microscope into a PCC 7942 flask culture or a flat-bottomed multi-well cell culture dish (Costar, Corning), which contains a dense culture of Streptococcus elongatus as a diluent, First isolate the herbivores and then cultivate the herbivores. Grazer cultures are cultured under low light conditions (10 μmol photons ⋅ m-2 ⋅ s-1) at 30°C or more conventionally at room temperature under ambient light conditions. Compared with healthy PCC 7942 cultures, the grazing cultures were yellow, resulting in visible clumps of the few remaining elongated S. elongatus cells, and they were mainly composed of ciliates when viewed under an optical microscope on a dissecting microscope. Using ITS1 (5'-AGG AGA AGT CGT AAC AAG GT-3') and ITS4 (5'-TCC TCC GCT TAT TGA TAT GC-3') primers, a ~600-bp band was generated. PCR was carried out in accordance with the Q5 standard protocol, the annealing temperature was 55 ℃, and the extension time was 3 min 45 s. The PCR products were purified using Zymo DNA Clean & Concentrator-5 kit, and sequenced by Sanger sequencing using ITS1 primers. The resulting sequence is analyzed by the basic partial comparison search tool to determine the closest identified species is Tetrahymena tropicalis, Tetrahymena thermophila or Tetrahymena thermophila.

Three 20-liter clear polycarbonate bottles (Nalgene) with custom-made inlet/outlet polypropylene caps and stainless steel diaphragm compression fittings, sealed with Viton O-rings, for gas-tight liquid and gas sampling, designed to prevent surrounding Harmful organisms in the environment enter the 18 L BG-11 and autoclave (SI appendix, Figure S1). After cooling, inoculate the large bottle with the 2 L S. elongatus PCC 7942 culture previously grown in the 2-L flask to an OD of 0.08 to 0.1 at 750 nm, so that the large bottle is inoculated to an OD of ~0.01 at 750 nm. Carboys are connected to the sampling equipment, as shown in Figure 1 and SI appendix, Figure S1. The culture was grown continuously under cold (4,100 K) fluorescent light, starting with 135 μE of incident radiation on the front of each large bottle, and increasing to 190 μE after 24 hours of growth. At this time, self-shading reduces the risk of photobleaching. At 8, 13, and 23 days after inoculation, Tetrahymena was added to Carboys 1, 2 and 3 at a calculated concentration of 0.1 cells/ml.

The clean zero air obtained from the generator (Sabio 1001) and adjusted by the mass flow controller (Alicat) to 3 standard liters per minute (SLPM) enters each tank through a 47-mm 0.2-µm PTFE filter, The air input line ends with a submerged circular length of 1/4 inch outer diameter perfluoroalkyl (PFA) tube, so that all input air is directly released into the culture in the form of bubbles. This structure distributes bubbles throughout the large bottle to fully mix and supply air to promote the growth of the culture. The liquid sample was collected from the large bottle using the three-way connection of the Tygon tube to the compression fitting on the lid and the one-way check valve manifold, all of which were connected to the large bottle for autoclaving during the preparation of the medium. The valve group allows the use of 70% ethanol to sterilize the liquid sample line, while significantly reducing the risk of backflow of ethanol or microorganisms into the bottle. The output headspace gas of each large bottle flows into a custom Labview-controlled four-channel solenoid valve array through a 3.175 mm inner diameter PFA tube, which switches between each large bottle and clean zero air within a 15-minute sampling period (Figure 1C). The sample air from the solenoid array is sucked in by the CIMS at a fixed flow rate of 1.8 SLPM, and the excess sample gas overflows to the downwardly directed exhaust port to discharge condensed water droplets (Figure 1B).

The ultra-high purity N2 gas produced by the evaporation of the liquid nitrogen dewar is bubbled through a 250 ml Pyrex bottle, which contains 2.2 SLPM of H2O with liquid chromatography mass spectrometry purity. The humidified N2 leaving the bottle enters the Po-210 alpha particle source (20 mCi) to generate protonated water clusters (H2O) nH+, where n = 1,2,3,..., the IMR ( PIMR = 23 Torr). The input flow of sample and reagent gas into the IMR is controlled by two online critical orifices (O'Keefe) of 1.8 SLPM. The chemical ionization of the analyte in the IMR by the protonated water cluster proceeds through several possible reaction mechanisms, which are controlled by the chemical energy of the analyte and reagent ions. These reactions are detailed in the equations. 1–4 (39), M+X+→M++X charge transfer[1] M+HX+→MH++X proton transfer[2] MH+X+→M++HX hydride abstraction[3] X++ M +Z→MX++Z adduct formation, [4]

Where M is the analyte, X is the reagent, and Z is the third object needed to take away excess energy. For (H2O)nH+, it is assumed that the main reaction channel is through proton transfer, which occurs when the proton affinity of the analyte exceeds that of water clusters. For (H2O)nH+, the proton affinity ranges of n = 0,1,2,3 water clusters are 691, 694, 730, and 769 kJ ⋅ mol−1; however, it is difficult to predict the true distribution of water clusters in IMR. Because the observed mass spectrum may not directly reflect this (48). Given the relatively high pressure of the IMR, it is expected that the water clusters will reach equilibrium cluster distribution much faster than their residence time in the IMR (~100 ms).

In chemical ionization, the humidity in the reaction chamber will greatly affect the distribution of ions and the kinetics of chemical ionization, which usually requires humidity calibration or internal standards to explain the changes in sample humidity (24). Although the humidity of the sampling line was not monitored in these experiments, the temperature of the sampling line was kept constant at 30 °C, while in the temperature-controlled laboratory, the temperature of the large bottle tank was 23 °C. In previous experiments that did not use pipe heating, the major challenge of condensation was encountered. From this, we assume that the pipe remains very close to the saturated humidity. In addition, the analysis of water cluster ions during one of the Carboy experiments (SI appendix, Figure S10) also showed that the ratio of water cluster ions to each other is very stable. Given the huge changes in humidity, we may see these ratios drift toward larger clusters when they are wetter, and toward smaller clusters if they are wetter.

After ionization, the reagent and product ions further enter the instrument, starting from the electrokinetic ion funnel (PIF = 0.3 Torr). This funnel uses radio frequency (RF) pseudopotentials to radially confine ions. The funnel is used in more than 30 concentrics with gradually decreasing diameters. Brass plate. The ions transmitted through the funnel enter the RF transmission-only quadrupole (PQuad = 5 × 10−5 Torr), which guides the ions to the final chamber: a commercial orthogonal extraction time-of-flight mass spectrometer (PToF = 4.0 × 10- 7 Torr) and V-shaped microchannel plate detector (Tofwerk AG) (Figure 1A). Approximately 60,000 mass spectra are averaged per second on the analog-to-digital converter and transferred to the computer for storage for later analysis. For the analysis in this work, the mass spectrum window was limited to between 10 and 400 m/z because of the lack of ions observed above this mass range. For the spectra obtained in this experiment, the mass resolution is ~1,200 full width at half maximum. The MS data is baselined, calibrated and analyzed in Tofware (Igor Pro-7 graphical user interface plug-in). The time series data of unit mass m/z is averaged into an interval of about 15 minutes to reduce the data density, and is separated by large bottles sampled using complementary mask files generated by the solenoid valve array. Approximately 5 minutes of MS data collected at the beginning of each time period are discarded to remove the signal that is equilibrating between the samples.

Under a dissecting microscope with a 25x objective lens and a 10x eyepiece, the Tetrahymena cell count was obtained by manually counting the ciliates present in 1 to 10 microliter droplets of the culture. Alternatively, for low-density herbivore cultures, the same dissecting microscope setup can be used to quickly scan 200 μL of culture in a 96-well plate. For all cell counts, at least five separate sample counts are collected and averaged to calculate the cell density of the culture or inoculum.

The cyanobacteria culture sample collected manually is analyzed in a transparent plastic cuvette with an OD of 750 nm, or a TECAN Infinite M200 plate reader is used to analyze the absorbance and fluorescence spectra of 200 µL samples in a 96-well plate to collect the OD750 value and The excitation wavelength of the autofluorescence value is 590 nm, and the emission wavelength is 670 nm.

By spotting a 5 µL sample on a BG-11 Omni plate (BG-11 with 0.04% glucose, 5% Luria broth, and 1.5% agar) and overnight at 30 °C (49). All S. elongatus flasks and large flask cultures were proven to be free of contaminants before adding Tetrahymena.

After each Carboy is inoculated with Tetrahymena grazer, connect a low-flow (~1 mL/min) peristaltic pump, and pump the Carboy liquid into a quartz 1-cm2 flow-through cuvette through a 1/8” outer diameter Tygon tube, and then return the liquid Return to the bottle in a closed loop. Place the cuvette in a custom-made fluorescence spectrometer that irradiates the cells with 420 nm light from a photodiode laser (Thorlabs). Orthogonal directional photoelectric with 650 nm high-pass filter The diode detector (Thorlabs) measures the fluorescence, and the Labview interface records the fluorescence at a rate of 1 Hz. After the experiment is completed, the continuous fluorescence data is subtracted from the background, normalized, and averaged to a 10-minute interval.

Fill a 10 mL crimp bottle with 4 mL of BG-11 growth medium, 4 mL of previously grown PCC 7942 culture with an OD of ~0.25, or 4 mL of PCC 7942 culture with ~370 Tetrahymena cells . During the experiment, the vial was crimped sealed and kept under a fluorescent light. For each microalgae-herbivore experiment, 10 vials were prepared, of which 5 were only algae controls, and 5 were infected with Ciliate Tetrahymena at the beginning of the experiment. On each day of the experiment, one untested vial of each group was sampled through SPME fiber, which has a 50-micron divinylbenzene/carboxyl/polydimethylsiloxane (solid gray, Supelco) stationary phase. The headspace of the vial was equilibrated for 24 hours. The samples were analyzed on Agilent Technologies 7820A/5975 GC/MS using Agilent BP5 column. The following GC/MS parameters are used to analyze SPME fibers: inlet temperature: 250 °C; splitless injection, carrier gas flow: 2 mL ⋅ min-1; GC temperature program: 35 °C for 3 minutes, then +7 °C ⋅ min-1 to 230 °C, hold for 2 minutes; MS transfer line temperature: 260 °C; MS scanning range: m/z 32.5 to m/z 300 for more than 1 second. The MS data uses a statistically sensitive nonlinear iterative peak clipping algorithm (50) to establish a baseline and analyzes it in OpenChrom (51) (https://lablicate.com/platform/openchrom). The MS spectra were identified using the NIST14 Electron Ionization Mass Spectrometry database. The results were then exported to Igor Pro-7 (Wavemetrics) for plotting.

Comma Separated Values ​​(CSV) and Raw Hierarchical Data Format (HDF) data have been deposited in the University of California San Diego Library, https://doi.org/10.6075/J0GH9GHW (52).

We thank Joseph Manson for assisting in the construction of the hardware and software of the valve block, and Joseph Mayer for assisting in the construction of custom bottle caps. We thank Kerry Kizer, Laura Lowe, and Paul Kasrazadeh for their help in preparing the culture for the experiments in this article. This material is based on work supported by the DE-EE0007094 grant from the Department of Energy.

↵1J.SS and RS have made equal contributions to this work.

Author contributions: research on the design of JSS, RS, KAP, and RSP; research on JSS, RS, ANM, LC, and SLS; analysis of data for JSS, RS, KAP, and RSP; JSS wrote this paper.

The author declares no competing interests.

This article is directly contributed by PNAS.

This article contains online support information https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2106882118/-/DCSupplemental.

This open access article is distributed under the Creative Commons Attribution-Non-Commercial-No Derivative License 4.0 (CC BY-NC-ND).

Thank you for your interest in advertising on PNAS.

Note: We only ask you to provide your email address so that the people you recommend the page to know that you want them to see it and that it is not spam. We do not capture any email addresses.

Feedback privacy/legal

Copyright © 2021 National Academy of Sciences. Online ISSN 1091-6490. PNAS is a partner of CHORUS, COPE, CrossRef, ORCID and Research4Life.