Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Dec 5.
Published in final edited form as: Analyst. 2023 Oct 5;148(20):5270–5278. doi: 10.1039/d3an01276j

Two-Dimensional Isomer Differentiation Using Liquid Chromatography-Tandem Mass Spectrometry with In-Source, Droplet-Based Derivatization

Derik R Heiss a,b, Enoch Amoah a, Abraham K Badu-Tawiah a
PMCID: PMC10696534  NIHMSID: NIHMS1944587  PMID: 37740330

Abstract

Saccharides are increasingly used as biomarkers and for therapeutic purposes. Their characterization is challenging due to their low ionization efficiencies and inherent structural heterogeneity. Here, we illustrate how the coupling of online droplet-based reaction, in a form of contained electrospray (ES) ion source, with liquid chromatography (LC) tandem mass spectrometry (MS/MS) allows the comprehensive characterization of sucrose isomers. We used the reaction between phenylboronic acid and cis-diols for on-the-fly derivatization of saccharides eluting from the LC column followed by in-situ MS/MS analysis, which afforded diagnostic fragment ions that enabled differentiation of species indistinguishable by chromatography or mass spectrometry alone. For example, chromatograms differing only by 2% in retention times were flagged to be different based on incompatible MS/MS fragmentation patterns. This orthogonal LC-contained-ES-MS/MS method was applied to confirm the presence of turanose, palatinose, maltulose, and maltose, which are structural isomers of sucrose, in three different honey samples. The reported workflow does not require modification to existing mass spectrometers, and the contained-ES platform itself acts both as the ion source and the reactor, all promising widespread application.

Introduction

This study showcases an online droplet-based phenyl-boronic acid reaction that has potential to make mass spectrometers selective toward saccharide isomers. When coupled with liquid chromatography (LC) tandem mass spectrometry (MS/MS), the accelerated online reaction system affords true orthogonal results to differentiate disaccharide isomers based on chromatographic retention and collision-induced dissociation (CID) MS/MS fragmentation of reaction products. In this case, orthogonal is used to indicate the independent data obtained from the two methods, the combination of which deliver insight on structural connectivity among disaccharide isomers. The method for post-column in-source droplet-based real-time analyte derivatization offers two advantages: (1) reduced band broadening since long mixers/reactors are not required – chemical reactions occur during the stages of droplet formation in electrospray ionization (ESI) and (2) more accurate standard selection for LC validation since the droplet reaction occurs after the (suspected) unmodified analytes are separated. That is, our method performs reactions (post-column) during the stages of ESI droplet formation (not before and not after). In this manner, our platform provides opportunities to obtain highly efficient data by optimizing the conditions for separation and ionization independently. For the first time, we apply the phenylboronic acid chemistry for disaccharide isomer differentiation, which is achieved in an online LC-MS/MS methodology.

Carbohydrates are a structurally complex class of molecules that serve diverse biological functions from building and maintaining energy stores to cellular signalling, immune response, and gene regulation.17 Physiological responses differ depending on the nature of the constituent monomers, the configuration of stereocenters and anomeric sites, and connectivity at the glycosidic linkage810, the varying combinations of which govern bioavailability and metabolism. For example, sucrose and palatinose are both disaccharides comprised of one glucose and one fructose monomer that differ only in the connectivity of the glycosidic bond. However, palatinose is half as sweet as sucrose and is metabolized much more slowly, making it a favorable replacement for those with diabetic or pre-diabetic tenden-cies.11

Mass spectrometry (MS) has proven to be a valuable tool for detection and structure elucidation of carbohydrates due to its versatility and high sensitivity.1019 Various LC-MS1923, gas chromatography-mass spectrometry (GC-MS)2426, matrix assisted laser desorption electrospray ionization-mass spectrometry (MALDI-MS)9,27, and ion mobility-mass spectrometry (IM-MS)8,12 methods have been reported. However, differentiation of isobaric saccharides is difficult using MS alone.

Tandem mass spectrometry provides a powerful means of revealing connectivity and conformation for structural studies and imparting additional specificity for targeted methods. Collision-induced dissociation has become the most commonly applied MS/MS technique for saccharide characterization due to its wide availability.10,16,18,20,21,28,29 More recently, specialized fragmentation techniques capable of generating abundant cross cleavages to elucidate branching and glycosidic bond location have also been utilized including ultraviolet and infrared photodissociation13,14,30 as well as electron capture dissociation15,31. However, challenges still exist in differentiating isomers, especially at low concentrations or when multiple isobaric species exist in a mixture. In these cases, an MS-independent separation technique such as LC can be helpful in providing important information to supplement MS/MS identification. Unfortunately, distinct fragment ions are typically not observed when metal adducts (Na, Ka, Li) of sugar isomers are subjected to collisional activation, which is the most common form of MS/MS that is available on almost all commercial mass spectrometers.32

Our approach makes CID selective to saccharide isomers without instrument modification. We combine LC-MS/MS analysis with online, microdroplet-based derivatization to both enhance sensitivity and generate more abundant and information-rich MS/MS spectra for isomer differentiation. Real-time derivatization is achieved during ionization using a coaxial flow contained-electrospray ion source (contained-ESI) previously developed in our lab.33,34 Conventional ESI-MS/MS analysis without derivatization offers poor sensitivity for saccharides due to low ionization efficiency.35 We previously demonstrated that droplet-based derivatization using phenylboronic acid generated much more abundant precursor and product ions in comparison with the underivatized saccharides, resulting in sensitivity enhancements of up to two orders of magnitude and improved capabilities for isomer differentiation.36

Chemical derivatization is a well-established strategy for overcoming the limitations of available instrumentation and methods. It is commonly used to expand detectability into adjacent areas in the chemical space, to drive signal enhancement for analytes that exhibit low sensitivity, and to improve selectivity in separations. However, conventional derivatization protocols can be resource intensive, complex, and time-consuming. Our overall objective is to exploit the virtues of derivatization while mitigating its shortcomings by shifting the procedure from the benchtop to the analytical platform. The ability to carry out derivatization reactions in seconds to milliseconds timescale during the ionization process is revolutionary. This is possible because of enhanced reaction kinetics inherent in desolvating electrosprayed microdroplets. The large surface area-to-volume ratio, increased reagent concentration and surface charge density during microdroplet desolvation, and enhanced mixing within the droplet caused by temperature, electric field, and concentration gradients all contribute to reaction rate enhancement.3749

For this work, the LC-contained-ESI-MS/MS platform with online, droplet-based phenylboronic acid derivatization was applied to identify the disaccharide isomeric contents in three different raw honey samples. Isomer identification is accomplished using retention time data, MS/MS ion profiles, and specific diagnostic fragment ions in some cases. Collectively, we were able to differentiate isomers that differed only by 2% in retention times.

Experimental

Contained-Electrospray Ion Source.

The contained-ESI source is an adapted version of an electrospray device that allows the infusion of liquid or vapor phase reagents directly into the ion source to modify ionization conditions or facilitate reactions. The sample (e.g., LC eluate) and reagent are introduced coaxially via separate capillaries that converge at the tip of the emitter. Mixing and reaction occur within the Taylor cone and ensuing microdroplets in parallel with ion formation; thus, high yield reactions can be achieved within the lifetime of electrosprayed droplets.34,37 In addition to improved saccharide analysis, the contained-ESI source has been used to study protein folding, mitigate matrix suppression, accelerate the enzymatic hydrolysis of lipids, and perform crosslinking experiments to study protein structure.33,34,38,50

A schematic of the coaxial contained-electrospray ion source configured for LC-MS analysis is shown in Figure 1 and images of the setup are included in Figure S1. The source is constructed using a stainless-steel Tee with compression fittings (Swagelok®, Solon, OH). Separate fused silica capillaries of 100 μm internal diameter each (190 μm outer diameter) deliver the LC eluate and derivatization reagent to the ion source. The two inner capillaries extend beyond a larger (outer) fused silica capillary of 450 μm internal diameter. As analytes elute from the column, they mix with the derivatization reagent within the Taylor cone and subsequent electrosprayed microdroplets.

Figure 1.

Figure 1.

A schematic representation of the dual-feed coaxial contained-electrospray ion source customized to facilitate in-source derivatization and ionization of analytes as they elute from an HPLC column.

The high voltage is applied to the stainless-steel needle delivering the reagent via alligator clip. After exiting the column, the LC eluate passes through a short (~30 cm) piece of black PEEK tubing (1/16” OD, 0.004” ID) which is connected to a grounded stainless-steel coupler to shield the HPLC electronics from the ion source. Nitrogen sheath gas is delivered to the ion source via polypropylene tubing and exits through the outer capillary concentric to the two inner capillaries. Sheath gas assists desolvation and spray stabilization and can be used to adjust reaction time by increasing or decreasing microdroplet lifetime.34

Chemicals and Reagents.

Maltulose monohydrate (99.3%) was purchased from Chem-Impex Intl. (Wood Dale, IL). Sucrose (99.5%), α-lactose monohydrate (99%), maltose monohydrate (99%), palatinose hydrate (>98%), and D-(+)-turanose (>98%) were purchased from Sigma-Aldrich (St. Louis, MO). Phenylboronic acid (PBA, >97%) was obtained from Strem Chemicals (Newburyport, MA). Acetonitrile (HPLC grade) and ammonium hydroxide (35% solution in water) were purchased from Fisher Scientific (Pittsburgh, PA). High purity deionized water with 18.2 MΩ·cm resistivity was prepared using a Milli-Q filtration system (Merck Millipore, Burlington, MA). High-purity nitrogen was used as the sheath gas.

Honey samples were all US Grade A and consisted of Nature Nate’s (McKinney, TX) Natural 100% Raw Unfiltered Honey (honey #1), locally sourced raw honey from Union County, OH (honey #2), and Kirkland’s Signature (Costco Wholesale, Seattle, WA) Organic Raw Honey (honey #3).

Preparation of Sugar Solutions, Reagent Solutions, and Honey Samples.

Individual stock standards for each sugar were prepared in water and used to formulate mixed standard solutions containing sucrose, turanose, palatinose, maltulose, maltose, and lactose at 10 μM each in 80:20 acetonitrile/water for analysis. The phenylboronic acid reagent was prepared by diluting the neat material in water to achieve a concentration of 8 mM. The pH of the solution was then adjusted to ~10 using ammonium hydroxide before being diluted two-fold with acetonitrile to produce a final reagent solution at 4 mM in 1:1 acetonitrile/water.

Approximately 500 mg of each honey sample was diluted with water to generate a 50 mg/mL solution. Five milliliters of each diluted honey sample were then transferred to a Microsep Advance 3k MW cutoff Centrifugal Filter (Pall Corp., Ann Arbor, MI) and centrifuged at ~3000 rpm for 60 minutes. The filtrates were diluted five-fold with acetonitrile and vortexed to mix prior to analysis.

Liquid Chromatography-Contained-Electrospray-Tandem Mass Spectrometry (LC-contained-ESI-MS/MS) Analysis.

Analyses were performed using an Agilent 1100 HPLC (Agilent Technologies, Santa Clara, CA) coupled to a Thermo Finnegan LTQ linear ion trap mass spectrometer (Thermo Scientific, San Jose, CA) operating in negative-ion mode. Data were acquired and processed using Thermo Fisher Scientific Xcalibur 2.2 SP1 software.

Chromatographic separation of the disaccharide mix was achieved using a Waters Acquity BEH Amide column of 100 mm length, 2.1 mm diameter, and 1.7 μm particle size (Waters Corp., Milford, MA), and an isocratic separation using a mobile phase of 84:16 (v:v) acetonitrile/water with 0.1% (v:v) ammonium hydroxide. The flow rate was 150 μL/min, and a 5 μL injection volume was used for all analyses.

Mass spectrometer inlet temperature and source voltage were optimized previously.36 Sheath gas pressure was increased slightly for this method compared to our previous, direct-infusion work in order to accommodate the larger flow rate from the LC. The final source conditions were as follows: sheath gas pressure (N2) = 20 psi, source voltage = −6.5 kV, capillary temperature = 325 °C, phenylboronic acid reagent concentration = 4 mM, and reagent flow rate = 5 μL/min. An optimal collision energy (CE) for the product ion scanning method in CID was established by infusing 10 μM solutions of each individual sugar into the contained-ESI source coaxially with the reagent and adjusting CE settings until the response for the base product ion was maximized. The final method conditions are shown in Table S1.

RESULTS AND DISCUSSION

In previous work, we demonstrated significant gains in sensitivity for saccharides using in-source, droplet-based phenylboronic acid derivatization (Scheme 1).36,51 This approach has been used to functionalize many classes of compounds including steroids, sugars, and carbohydrates5260 and is optimal for the current application due to the aqueous stability of the reagent and the relatively mild conditions required for reaction. Further, because this reaction targets diol moieties, it can be applied for both reducing and non-reducing sugars.

Scheme 1.

Scheme 1.

General reaction of phenylboronic acid with vic-diols. The phenylboronic acid was prepared in a final solution of acetonitrile/water (1:1, v/v) at 4 mM concentration, containing ammonium hydroxide (pH ~10). The sugar was eluted from LC column.

LC-contained-ESI-MS/MS Method Development for Sucrose Isomers.

Differentiation of isomers can be challenging using mass spectrometry alone. In many cases, the compounds of interest generate the same precursor and product ions making it difficult to determine which isomer is present or if the sample is a mixture of isomers. Chromatography can be helpful, although at times this also proves inadequate if sufficient resolution cannot be attained. For this work, we integrated the two orthogonal techniques to allow the differentiation of isomeric disaccharides in complex mixtures. Again, the novel aspect of the work is related to the integration of on-the-fly chemical reaction, that enables characteristic MS/MS spectra to be obtained for isobaric sugars to yield useful orthogonal data when coupled with LC. The result is an efficient separation method that is coupled with tandem mass spectrometry operating in product ion scanning mode. At the interface is the contained-ESI source, which serves to generate the requisite ions for mass analysis while simultaneously performing in-source, droplet-based derivatization to boost sensitivity and facilitate abundant and distinct MS/MS spectra for the disaccharide isomers.

The LC method relies on HILIC conditions for chromatographic separation, which has been shown to be well suited for sugars due to their relatively polar nature.6163 We employed an amide stationary phase using mobile phases consisting of acetonitrile and water. Methanol was avoided due to its tendency to react with the derivatization reagent and the open-chain tautomers of reducing saccharides. Ammonium hydroxide was added to the mobile phases as a modifier to enhance retention, improve ionization conditions, and minimize peak broadening/splitting that can occur for reducing sugars as a result of anomeric mutarotation. The increased pH also provides ideal conditions for the derivatization reaction to occur.36,52

In previous work designed to maximize sensitivity gains using in-source derivatization, we found that reaction yield decreased as LC flow rate increased due to the production of larger electrospray droplets.50 Thus, we intentionally limited the LC flow rate to 75 μL/min by employing a microbore column. However, since the primary driver of the current application is differentiating isomers rather than maximizing sensitivity, we were able to operate at higher LC flow rates, allowing us to decrease analysis time and enabling the use of conventional LC columns. Five different flow rates were tested ranging from 75 μL/min to 175 μL/min to evaluate the effect of flow rate on isomer resolution and analysis time (see Figure S2). In addition, reaction yield (ratio of peak areas for the derivatized species and the underivatized species) was measured at each flow rate to assess the impact of LC flow rate on the droplet reaction. Not surprisingly, we found that reaction yield decreased as flow rate increased (see Figure S3). The decline in yield was not as dramatic as expected, and in fact most of the sensitivity gains afforded by the droplet reaction were preserved even at flow rates as high as 175 μL/min. Ultimately, to balance chromatographic resolution with analysis time and reaction yield, we selected a flow rate of 150 μL/min for the method. Although the flow rate is high enough for standard bore columns, it is still low enough to allow the use of the 1.7 μm particle sizes typically reserved for ultra-high performance liquid chromatography (UHPLC) without exceeding pressure limitations of current HPLC equipment. In this way, we can maximize resolution and peak shape and reduce analysis time.

The six disaccharide isomers used in this work are shown in Figure 2. Four of the compounds, sucrose, turanose, palatinose, and maltulose, are constitutional isomers composed of glucose and fructose monomers differing only in the connectivity of their glycosidic linkage. The remaining two compounds, maltose and lactose, are structural isomers containing glucose-glucose and galactose-glucose monomer combinations, respectively.

Figure 2.

Figure 2.

Six isomeric disaccharides included in the method.

LC method development involved testing various mobile phases including acetone/water and acetonitrile/water solvent mixtures. Acetone was tested due to its relatively similar physical properties to acetonitrile (e.g.; viscosity, volatility) making it a suitable solvent for LC-MS experiments.64 Mobile phases containing acetone offered slightly improved retention in comparison with acetonitrile but resulted in broad peaks for maltose and lactose due to anomeric mutarotation, even in the presence of 0.1% NH4OH and using elevated column temperatures (80 °C). Acetonitrile-containing mobile phases produced better peak shapes overall, especially for maltose and lactose. Isocratic conditions were found to be most favorable for resolving the four constitutional isomers, since they are the most closely related in structure.

Previous work demonstrated that sucrose and lactose form primarily the monosubstituted phenylboronate ester during in-source derivatization with phenylboronic acid, which is detected as deprotonated species at m/z 427. Unsurprisingly, we also found this to be the case for the other four isomers: palatinose, maltulose, maltose, and turanose. In addition to the monosubstituted precursor ions (M/PBA) at m/z 427, we also observed significant formation of the hydroxylated species (M/PBA+OH) at m/z 445 in the negative-ion mode, as shown in Figure S4. Interestingly, the ratio of the deprotonated and hydroxylated precursor ions varied among the different isomers. CID of the hydroxylated precursor ions provided little value, resulting mainly in the formation of the deprotonated product ion. Thus, the product ion scan used for the new method was based on the deprotonated precursor ion of the monosubstituted MS/MS analysis of the deprotonated, monosubstituted derivatives of the six sugars generated unique product ion spectra (see Figure S5). The primary fragment ions generated by all of the isomers were m/z 229, m/z 247, and m/z 265 representing the loss of C6H14O7, C6H12O6 and C6H10O5, respectively. A loss of water was also observed in all cases at m/z 409. However, there were a few diagnostic product ions produced by only one or two of the isomers that could prove very useful for isomer discrimination. For example, m/z 349 was only generated for lactose and maltose while m/z 391 and 337 were only observed for palatinose and turanose. In the case of lactose and maltose, m/z 427 first losses C2H4O2 (MW 60 Da) to form m/z 367. Subsequent loss of H2O leads to the formation of m/z 349. Formation of the diagnostic ion m/z 391 can be attributed to loss of H2O from m/z 409. Similarly, loss of CO (MW 28 Da) and CO2 (MW 44 Da) from m/z 409 affords the diagnostic ion at m/z 337. Other ions were observed that were unique to only a single isomer, including m/z 305 for palatinose, and m/z 221 and m/z 263 for maltose.

Although the sugars tested generated mostly the same product ions, the relative ratios of these ions differed among the isomers (Figure S6). Even in the absence of unique ions, these ion ratio profiles can be used to differentiate isomers based solely on MS/MS analysis after PBA derivatization. Coupling this MS/MS data with retention time information from LC in our LC-contained-ESI-MS/MS method provides further validation. The optimized LC-contained-ESI-MS/MS method for the orthogonal analysis of isomeric disaccharides is summarized in Table S1. A chromatogram representing a mixture of the six sugars is shown in Figure 3Ai. Using the isocratic HILIC LC conditions, nearly baseline resolution was obtained among the four constitutional isomers (sucrose, turanose, palatinose, maltulose). Maltulose and maltose could not be fully resolved, and peak broadening was still evident for maltose and lactose due to the anomeric mixture of each peak. However, the method still offers sufficient chromatographic separation of the isomers to allow mixture analysis. Further, although sensitivity enhancement was not the primary objective of this study, significant gains were still evident using in-source PBA derivatization even at the higher LC flow rate (150 μL/min), as only small indications for two of the six sugars (sucrose, turanose) were observed in the chromatogram for the underivatized analytes (Figure 3Aii). Proper optimization could enable detection of the underivatized sugars as other ion types (e.g., Cl adduct), but sensitivity was not the focus on this work. In addition to the chromatographic resolution, unique product ion spectra were collected for each of the six sugars during the analysis (Figure 3B), thus adding another, orthogonal dimension for isomer distinction.

Figure 3.

Figure 3.

(A) Chromatograms generated using LC-MS/MS product ion scanning (i) with in-source PBA derivatization and (ii) without derivatization. (B) Negative-ion mode product ion spectra for each of the six sugars with PBA attributed to the numbered peaks in the standard chromatogram (Ai), all at m/z 427. MS/MS isomer differentiation is based on ion profiles. See Figure S5 discussion on diagnostic fragment ion detected for some isomers.

Analysis of Sucrose Isomers in Honey.

Sucrose is ubiquitous in the natural world. Its wide availability and relative ease of production makes it the most prevalent sweetener in the food industry.65 However, excess sucrose consumption can lead to health problems including tooth decay, obesity, high blood pressure, cardiovascular disease, and type 2 diabetes. Consequently, there is a need to identify, characterize, and market alternative sweeteners that have derivative (m/z 427). To avoid distribution of saccharide signal into monosubstituted precursor (M/PBA) and hydroxylated species (M/PBA+OH), we are exploring other reactions in parallel studies.65 the same effect on the consumer palette but mitigate the health challenges associated with high sugar intake.

Honey, a complex mixture of proteins, carbohydrates, amino acids, organic acids, and other small molecules, has long been used as a sweetener in foods and drinks. Its saccharide profile consists mainly of fructose (38%) and glucose (31%), but also contains various disaccharides (~9%) including sucrose, maltose, turanose, palatinose, maltulose, trehalulose, nigerose, and gentiobiose.66,67 Some of these disaccharides have been identified as potential alternative sweeteners that not only reduce the ill effects of sugar consumption but can also provide biological value beyond energy production. For example, turanose has been shown to reduce inflammation and inhibit lipid accumulation and may be useful for controlling obesity, while palatinose has prebiotic activity that can aid in managing chronic diseases.68,69 Although these isomers provide only half of the sweetness of sucrose, they hydrolyze much more slowly both orally and gastrointestinally.70 Thus, they are less destructive to teeth and take longer to digest, leading to reduced blood sugar spikes. Characterization of saccharide profiles in honey has been accomplished using several techniques including liquid chromatography with ultraviolet absorbance detection (LC-UV)71 or amperometric detection72, ion chromatography-mass spectrometry73, and surface-assisted laser desorption/ionization mass spectrometry (SALDI-MS).67 Gas chromatography coupled to mass spectrometry (GC-MS) or flame ionization detection (GC-FID) have been used extensively but require offline derivatization prior to analysis.7275 These methodologies have certain advantages, but also present additional challenges with regard to complexity and limited specificity.

We utilized the new LC-contained-ESI-MS/MS method for the identification of select isomeric disaccharides in three different sources of honey. Chromatograms obtained from the analysis of honey samples are shown in Figure 4A. A chromatogram for a standard mix of the six sucrose isomers is also included for comparison. The overall disaccharide profiles are relatively similar among the three honey samples. Based on retention times and product ion spectra (Figure 4B), four of the sucrose isomers known to be present in honey were clearly identified (turanose, palatinose, maltulose, maltose). The presence of lactose in honey has been reported in some analyses, but not in others.7376 Lactose was not detected here. Sucrose is known to be present in honey at low concentrations but could not be clearly differentiated in this analysis. The peak at approximately 19.9 minutes present in each of the honey samples might be mistaken for sucrose based on retention time alone since there is only a 2% retention time difference in relation to the standard. However, the MS/MS spectrum for the unknown peak (Figure 5B) is clearly different from that of standard sucrose (Figure 5A), most notably in the presence of the m/z 391 product ion observed for the unidentified sugar which is not generated by sucrose. There may be a small contribution from endogenous sucrose, but this peak is clearly related to another, unknown disaccharide. This case illustrates the importance of the two-dimensional approach and the need for multiple, orthogonal modes of identification; deficiencies in the separation can be overcome by the mass spectrometer, and vice versa. Figure S7 shows the product ion spectra acquired for the peaks representing other disaccharides that could not be identified in this analysis. Other disaccharides reported to be present in honey but not evaluated in this work include trehalose, trehalulose, nigerose, and gentiobiose.7175 The addition of more disaccharide standards to the analysis might shed light on some of the unknown peaks present in the chromatograms.

Figure 4.

Figure 4.

(A) Chromatograms acquired for various honey samples in comparison with a standard mix of the six sucrose isomers tested. Analysis was performed using LC-contained-ESI-MS/MS with in-source PBA derivatization, with product ions detected in the negative-ion mode. (B) Negative-ion mode MS/MS product ion spectra for four sucrose isomers (after reaction with PBA) identified in the honey samples, all at m/z 427.

Figure 5.

Figure 5.

Typical negative-ion mode MS/MS spectra (at m/z 427) of monosubstituted phenylboronate esters of (A) standard sucrose and (B) unknown species in honey sample, as obtained from Figure 4A.

Conclusions

In this work, chromatographic separation of several isobaric sucrose isomers was used to supplement the characteristic product ion spectra generated after droplet-based PBA derivatization to provide a truly two-dimensional technique for isomer differentiation.

This inherently orthogonal manner of isomer distinction could be useful for the characterization of complex samples for a wide variety of compound classes. Leveraging the kinetic gains of droplet-based reactions via in-source derivatization has the potential to replace the complicated and costly benchtop derivatization model to improve chemical analysis in a variety of ways including better sensitivity, enhanced isomer distinction, and improved unknown identifications through structure elucidation.

Supplementary Material

Supporting Information

Acknowledgements

This work was supported by the U.S National Institute of General Medical Sciences (Award number R01GM149080) and Internal Research & Design (IR&D) funding from Battelle Memorial Institute

Footnotes

Conflicts of interest

The authors declare no competing financial interests.

Electronic Supplementary Information (ESI) available(Figure S1) Images of the LC-contained-ESI-MS setup and contained-ESI source; (Table S1) optimized LC-contained-ESI-MS/MS method parameters; (Figure S2) chromatograms at various flow rates; (Figure S3) effect of LC flow rate on derivatization reaction yield; (Figures S4 and S5) full scan and MS/MS spectra for PBA-derivatized sugars; (Figure S6) plot of product ion abundance ratios for the test sugars; (Figure S7) MS/MS spectra corresponding to four unknown peaks observed in honey chromatograms (PDF)

References

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

RESOURCES