Nanopore analysis of cis-diols in fruits

Fan, Pingping; Cao, Zhenyuan; Zhang, Shanyu; Wang, Yuqin; Xiao, Yunqi; Jia, Wendong; Zhang, Panke; Huang, Shuo

doi:10.1038/s41467-024-46303-x

Download PDF

Article
Open access
Published: 05 March 2024

Nanopore analysis of cis-diols in fruits

Nature Communications volume 15, Article number: 1969 (2024) Cite this article

3997 Accesses
14 Altmetric
Metrics details

Subjects

Abstract

Natural fruits contain a large variety of cis-diols. However, due to the lack of a high-resolution sensor that can simultaneously identify all cis-diols without a need of complex sample pretreatment, direct and rapid analysis of fruits in a hand-held device has never been previously reported. Nanopore, a versatile single molecule sensor, can be specially engineered to perform this task. A hetero-octameric Mycobacterium smegmatis porin A (MspA) nanopore modified with a sole phenylboronic acid (PBA) adapter is prepared. This engineered MspA accurately recognizes 1,2-diphenols, alditols, α-hydroxy acids and saccharides in prune, grape, lemon, different varieties of kiwifruits and commercial juice products. Assisted with a custom machine learning program, an accuracy of 99.3% is reported and the sample pretreatment is significantly simplified. Enantiomers such as DL-malic acids can also be directly identified, enabling sensing of synthetic food additives. Though demonstrated with fruits, these results suggest wide applications of nanopore in food and drug administration uses.

Nanopore analysis of salvianolic acids in herbal medicines

Article Open access 05 March 2024

Detection of exogenous sugars in pineapple juice using compound-specific stable hydrogen isotope analysis

Article Open access 14 April 2021

Evaluation of ambient mass spectrometry tools for assessing inherent postharvest pepper quality

Article Open access 01 July 2021

Introduction

Natural edible substances, including fruits, tea, honey, vegetables and many others contain rich cis-diols, such as 1,2-diphenols, alditols, α-hydroxy acids and saccharides^1,2,3,4. A thorough understanding of these compounds in food is essential for nutritional or healthcare purposes^5,6,7. Rapid and accurate analysis of food is also used for safety and quality control during food manufacture^8,9. However, these cis-diols from natural sources are invariably a mixture of compounds, which poses a technical challenge for simultaneous analyte identification using a single sensor and under the same condition¹⁰. Conventionally, high-performance liquid chromatography (HPLC) has been widely applied to the analysis of cis-diols in fruits¹⁰, tea^11,12 honey¹³ or vegetable¹⁴, but the chromatographic separation of such mixtures is complex and time consuming and requires equipment that is far from portable. In HPLC analysis, simultaneous detection of a variety of different analytes using the same chromatographic condition is also challenging^15,16,17. Other reported analytical methods applicable to cis-diols include spectrophotometry¹⁸, liquid chromatography tandem mass spectrometry (LC-MS/MS)^19,20 and gas chromatography (GC)^21,22. The resolution of spectrophotometry is however insufficient for accurate characterization of a variety of cis-diols. GC requires complex and poorly reproducible derivatization operations and LC-MS/MS methods require complex sample separation prior to the measurement.

Nanopore, which was originally developed for single molecule sequencing of nucleic acids²³, is a highly versatile sensor^{24,25,26,27,28,29,30,31,32,33}. When properly engineered to contain a specific reactive adapter in its lumen, a nanopore immediately becomes reactive and is sensitive to a set of corresponding small molecule analytes^34,35,36, which can react reversibly with the adapter to produce successive events. By producing highly characteristic event features for different small molecules, a nanopore can discriminate between different analytes that may be only slightly different in structure^32,37,38. This sensing capacity is thus suitable for direct and simultaneous analysis of a complex mixture of molecules (Fig. 1).

**Fig. 1: Nanopore analysis of complex samples.**

To use a specific reactive nanopore sensor, any sample can be decomposed into different categories of molecules, including target analytes, interfering analytes and non-interfering analytes (Fig. 1a). Target analytes are molecules of interest that can react with the nanopore to produce corresponding events (Fig. 1b). Simultaneous presence of multiple types of target analytes can be technically tolerated, as long as these target analytes can be fully distinguished by their nanopore event characteristics (Fig. 1c)^32,37,38. A multi-feature nanopore spectrum, which simultaneously considers multiple event features in a single plot, can be generated and used as a unique signature for a specific type of analyte (Fig. 1d). Meanwhile, the interfering analytes are molecules that are of no interest but can however react with the pore and can also produce nanopore events. However, as long as these events can be fully distinguished from that of all target analytes, their interference in the measurement becomes negligible and these events can subsequently be removed computationally (Fig. 1e). Non-interfering analytes are all molecules that fail to react with the nanopore and thus they don’t produce any events. Even when present in a large quantity, the non-interfering analytes will never appear as nanopore events and their impact can be ignored.

In this work, an MspA nanopore modified with a single phenylboronic acid is prepared. During single channel recording, it serves to reversibly react with cis-diols such as 1,2-diphenols, alditols, α-hydroxy acids and saccharides so that highly characteristic nanopore events are produced. Assisted by custom machine learning algorithms, a variety of cis-diols in natural fruits can be accurately and simultaneously identified. This sensing capacity also allows for high-resolution discrimination of enantiomeric DL-malic acid, demonstrating immediate applications in the detection of food additives in commercial fruit products. This technique could also be integrated into a miniaturized device, demonstrating its potential for portable sensing.

Results

Nanopore signatures of model cis-diols

Fruits, which contain a mixture of cis-diols such as 1,2-diphenols, alditols, α-hydroxy acids and saccharides^39,40,41, were used as an example for this demonstration. These cis-diols react reversibly with phenylboronic acid (PBA) in an aqueous environment. Ten types of cis-diols containing compounds including catechin (CAT), neochlorogenic acid (3-CQA), D-sorbitol (D-SOR), xylitol (XYL), L-malic acid (L-MA), L-tartaric acid (L-TA), citric acid (CA), (2R,3S)-isocitric acid (ICIT), D-glucose (D-GLC) and D-fructose (D-FRU) were selected as model analytes (Fig. 2a). They represent 1,2-diphenols, alditols, α-hydroxy acids and saccharides that are widely encountered in fruits^39,40,41. MspA-PBA, a hetero-octameric MspA nanopore modified with a sole phenylboronic acid (PBA) reactive adapter at site 90 (Fig. 2b), was prepared as previously reported^36,42,43,44 (Methods). The PBA at the pore constriction reacts reversibly with cis-diols to form a boronate ester⁴⁵, and specifically binds molecules containing cis-diols in a heterogenous mixture (Supplementary Fig. 1). At room temperature in an aqueous measurement environment, the capture and release of a cis-diol analyte by MspA-PBA can last for more than a few milliseconds, which is ideal for data acquisition by single channel recording. The rate of the reaction can also be finely tuned by changes in the environmental pH and temperature^45,46,47. The MspA nanopore, which has a conical lumen geometry that focuses the flow of ionic current to the pore constriction, also provides a high resolution of sensing in the discrimination of structurally similar compounds which has been well demonstrated in nanopore sensing of isomers of saccharides³⁶, alditols⁴³ and ribonucleotides^42,44. This high resolution thus enables direct recognition of target molecules from interfering analytes and accordingly, the need for sample separation is minimized.

**Fig. 2: Sensing of target *cis*-diols.**

Unless stated otherwise, all nanopore measurements were carried out in a custom nanopore device which has two chambers separated by a Teflon film containing an aperture of ~100 µm diameter. By convention, the chamber that is electrically grounded is defined as cis and its opposing chamber is defined as trans. After spontaneously forming a lipid bilayer on the aperture and with MspA-PBA inserted in this bilayer, all nanopore measurements were performed in a buffer of 1.5 M KCl, 100 mM MOPS, pH 7.0 and a continually applied +160 mV bias (Methods). A higher salt concentration and a higher applied bias generally produce nanopore events of a larger amplitude so that a better discrimination resolution is achieved (Supplementary Figs 2, 3). However, a too high applied bias may also generate a high risk of bilayer rupture. A high salt concentration may also lead to introduction of noises. Thus, a 1.5 M KCl buffer and a + 160 mV bias were found to be optimum.

To record the event characteristics of model compounds, each type of cis-diol was added as a sole analyte to both cis and trans chambers with a defined final concentration. All ten model analytes reported fully distinguishable nanopore events (Fig. 2c–l). These events are highly consistent when the same analyte was tested in measurements with different pores. However, their event characteristics are visually different when different compounds were tested. The consistency of event features for each analyte type was also verified by three independent measurements (N = 3) performed for each condition (Supplementary Figs 4–13). This further confirms that the high resolution of MspA can well discriminate between different analytes. The nanopore signatures of catechin (CAT), neochlorogenic acid (3-CQA), L-malic acid (L-MA), L-tartaric acid (L-TA), citric acid (CA) and (2R,3S)-isocitric acid (ICIT) have not been previously reported.

To quantitatively describe sensing events, event parameters including open pore current (I₀), residual current (I_b), blockage amplitude (ΔI), standard deviation (std), kurtosis (kurt), skewness (skew), dwell time (time) and median (med) were used and are defined in Fig. 1c. Results of the concentration dependence assay acquired from all ten model cis-diols were summarized in Supplementary Figs 14–23 and Supplementary Table 1. Based on the mechanism of nanopore sensing by single molecule reaction, simultaneous capturing of two cis-diols by an MspA-PBA is impossible. Thus, the event features of a specific analyte are independent of the analyte concentration. However, their rate of event appearance is highly concentration dependent and can be used for analyte quantification. The blockage amplitude is defined as ΔI = I_b − I₀. 1,2-diphenols, alditols and saccharides all appear as negative-going events (ΔI < 0, Supplementary Fig. 24) but all α-hydroxy acids report positive-going events (ΔI > 0, Supplementary Fig. 24). The phenomenon of positive-going events had been previously observed⁴⁶. The MspA nanopore is extremely sensitive to the presence of additive charges at its pore constriction^48,49. For instance, the WT MspA, which has negatively charged aspartic acid at its pore constriction, reports a much higher channel conductance than its mutant having asparagine introduced by mutagenesis. The negative charge of α-hydroxy acids should contribute to the increase of the ionic current, but the size of the small molecule analytes fails to provide any significant contribution to the blockage of the ionic current. Thus, binding of α-hydroxy acids generally reports positive-going events. On the other hand, although 1,2-diphenols, alditols and α-hydroxy acids all report a single type of nanopore event, the saccharides instead reported several event types (Fig. 2k, l). This may result from interconversion between their pyranose and furanose forms in an aqueous solution^36,47, as was observed in our previous study³⁶.

Although the event characteristics of different analytes are visually distinct from one another in a 2D event scatter plot of ΔI/I₀ and std, their event distributions still slightly overlap (Supplementary Fig. 25). However, when more event features were simultaneously considered, the accuracy of event identification was immediately improved. This is more clearly demonstrated in the multi-feature nanopore spectrums, in which the average means and standard deviations of five event features, including ΔI/I₀, std, kurt, skew and time are shown in the same plot (Supplementary Fig. 26, Supplementary Table 2). If necessary, more event features can also be included to further improve the event discrimination.

Event identification by machine learning

To enable automatic and unbiased event identification based on results from the five-feature nanopore spectra, a custom machine learning algorithm was developed. All machine learning was performed in the Classification Learner toolbox of MATLAB. Five event features, including ΔI/I₀, std, kurt, skew and time (Fig. 3a) were extracted to form a feature matrix. The label of each event was assigned as the type of the sole analyte used for data acquisition. The event features of a total of 5600 events from all ten types of model analytes were collected to form the training set which was used for model training (Methods). In the parallel coordinate plot, all five features contribute to the discrimination between events (Fig. 3b). According to the validation accuracies produced by different classification models, the Bagging Trees model reported the highest accuracy of 99.3% (Fig. 3c, Supplementary Table 3).

To test the generality of the model, an additional 1120 events, which didn’t participate in the model training, were used as a testing set. The results of event prediction for the testing set are summarized in Supplementary Table 3, in which the testing accuracy of most models was >80%. During testing, the Bagging Trees model again reported the highest accuracy of 99.1%. Thus, if not otherwise stated, the previously trained Bagging Trees model was selected as the optimum model for all future event predictions. The confusion matrix plot of the testing set generated by prediction of the Bagging Trees model is shown in Fig. 3d. The learning curves are also shown, from which the model is confirmed to be not overfitting and an input training data of 1000 is enough to gain a >98% prediction accuracy (Fig. 3e). The machine learning algorithm, which simultaneously considers five event features in the event identification, is shown to be accurate, efficient and unbiased. Ten cis-diols were also simultaneously measured using MspA-PBA in a 1.5 M KCl buffer with a continually applied +160 mV bias. Representative traces acquired at this condition were presented in Supplementary Fig. 27. All events were identified by the trained Bagging Trees model and the events on the trace were labelled accordingly. The resolution of the pore and the performance of the machine learning model are thus well approved.

Rapid analysis of natural fruit juice by a nanopore

The nanopore event characteristics of model analytes and the corresponding machine learning algorithm were used in the rapid analysis of cis-diol components in natural fruit juice. With prunes as the first example, a work flow of nanopore analysis of prune juice is summarized in Fig. 4a. Briefly, prunes were first pureed in a food processor and then centrifuged at 1900 g for 10 min to separate the solid material from the juice. To further remove solid residues, the juice was placed in an ultracentrifugation tube with a 3 kDa molecular weight cut off (MWCO) and then centrifuged at 2350 g for 10 min. The filtrate was collected and directly added to the nanopore device (Methods). Subsequently, the corresponding nanopore events were observed immediately (Fig. 4b). The results of nanopore measurements with natural fruit juice also report background noise events, appearing as non-clustered distributions in the scatter plot (Supplementary Fig. 28). To remove interference from them, a cluster analysis algorithm based on Density-Based Spatial Clustering of Applications with Noise (DBSCAN)⁵⁰ was introduced to computationally and automatically remove these interfering events (Supplementary Fig. 28). Subsequently, the trained Bagging Trees model was used to identify nanopore events in prune juice, enabling identification of 3-CQA, D-SOR, L-MA, D-GLC and D-FRU (Fig. 5b, c, Supplementary Movie 1). Three independent trails were performed at the same measurement condition and highly consistent results were shown (Supplementary Fig. 29). The high resolution of event discrimination by nanopore further guarantees that each type of model compound can be recognized reliably. These sensing features of nanopore enable direct measurement of fruit samples and without complicated sample separation (Fig. 4a). Instead, when the same juice was sensed by M2 MspA⁴⁸, which contains no PBA adapter, no nanopore events with well-defined features were observed (Supplementary Fig. 30). This further confirms that an unmodified MspA pore fails to provide any useful information for analysis of fruit juice and that the PBA modification is critical in the generation of cis-diol events.

**Fig. 4: Rapid analysis of natural prune juice.**

**Fig. 5: Rapid analysis of natural fruit juice.**

The operations demonstrated above were also similarly used with grape or lemon juice. Following corresponding nanopore measurements using MspA-PBA (Supplementary Figs 31, 32), background noise reduction with DBSCAN (Supplementary Figs 33, 34) and machine learning based event identification by the Bagging Trees model, the cis-diol components in grape and lemon juice were all accurately recognized. For grapes, the cis-diols identified were L-MA, L-TA, D-GLC, D-FRU (Fig. 5d, e, Supplementary Fig. 35 and Supplementary Movie 2), which are consistent with the results obtained by HPLC⁵¹ and for lemon, the identified events were from L-MA, CA, ICIT, D-GLC and D-FRU (Fig. 5f, g, Supplementary Fig. 36 and Supplementary Movie 3). Quantification of relevant cis-diols in above discussed fruit juices were also performed as described in Methods. The corresponding results were also presented in Supplementary Tables 4, 5, according to which grape has the highest D-fructose content and lemon has the highest CA content, respectively.

According to these results, L-MA, D-FRU and D-GLC were commonly observed from all three types of fruits but D-SOR and 3-CQA were only detected in prunes. It has been reported that D-SOR intolerance is common in humans and can often cause diarrhea⁵². Therefore, the prune is a natural laxative which can be used to treat constipation. 3-CQA on the other hand was reported to have antioxidant and anti-inflammatory effects⁵³. Therefore, proper consumption of prunes can prevent chronic inflammation. Among these three types of fruits, L-TA is only detected in grapes. It has been reported that the tartaric acid not only enhances the astringency of wine⁵⁴, but also affects its rheological properties and color⁵⁵. Thus, accurate sensing of L-TA by the nanopore may provide a rapid but accurate way to grade grapes in the wine industry. On the other hand, CA and its isomer ICIT were simultaneously detected in lemons, demonstrating the exceptional resolution of MspA in the discrimination of isomers from a complex sensing environment. Lemon is widely applied in the seasoning of food and drinks in daily life and these two acids make a strong contribution to the generation of the unique sour taste of lemon^56,57.

Though the use of the mini food processor and the centrifuge guarantees the efficiency and the reproducibility of juice generation, these operations are optional. To further simplify the workflow, the fruit can be manually squeezed and the generated juice can be directly analyzed by nanopore. To demonstrate this, manually squeezed lemon juice was ultrafiltered (3 kDa MWCO) at 3380 g for 3 min at 4 °C. The collected lemon juice was directly added to the nanopore sensor (Supplementary Fig. 37a, Methods). The whole operation took only 8 min and the measurement results (Supplementary Fig. 37b) are generally consistent with those demonstrated in Fig. 5f.

By taking grape juice as a demonstrative sample, the stability and consistency of this technique was also evaluated in a time-extended measurement (Supplementary Fig. 38). According to this demonstration, the nanopore assay could be continuously run for a few hours and a high consistency of results is shown. The characteristic event features acquired at the beginning of the measurements and a few hours later also appear completely identical (Supplementary Figs 38, 39). Though the MspA nanopore is protein in nature, according to previous investigations^58,59,60, it demonstrates a high rigidity and durability in the structure, which is ideal for long term measurement and device storage.

Rapid analysis of different varieties of kiwifruits

The capacity of MspA-PBA to discriminate between a large variety of cis-diols also suggests its use to distinguish between fruits of different varieties. To demonstrate that, three varieties of commercially available kiwifruits from Zespri^TM (New Zeeland), including Green, Sungold and Rubyred kiwifruit (Fig. 6a, d, g), were analyzed by nanopore. The juice generation was carried out as described in Methods. The measurement was performed using MspA-PBA in a 1.5 M KCl buffer and a + 160 mV bias was continually applied. To initiate the measurement, 2 μL kiwifruit juice was respectively added to both cis and trans chambers. Immediately afterwards, corresponding events were observed (Supplementary Figs 40–42). All raw data acquired from different trials of measurements were shown in Supplementary Figs 43–45. The raw data were further treated by outlier analysis using One-Class SVM (Supplementary Figs 46–48), which is an unsupervised machine learning algorithm to detect outlier events. Details of One-Class SVM model establishment is summarized in Methods. After outlier analysis, all data were grouped into inlier and outlier events. All inlier events were predicted using the previously trained Bagging Trees model. Whereas, all outlier events were further treated by cluster analysis using DBSCAN to detect the appearance of newly appearing event clusters.

Specifically, new clusters of events, which don’t belong to any standard events recorded in the machine learning database, were detected in Sungold kiwifruit (Supplementary Fig. 47) and Rubyred kiwifruit (Supplementary Fig. 48), suggesting that previously uninvestigated cis-diols were detected in these kiwifruit varieties. According to that reported in literatures^61,62, sucrose, which is a cis-diol not previously investigated by MspA-PBA, is also present in Sungold kiwifruit. It is thus speculated that the newly detected event cluster in Sungold kiwifruit might be from sucrose. To verify this hypothesis, sucrose was separately measured using MspA-PBA and its event features were recorded and confirmed to be consistent with that detected in Sungold kiwifruit (Supplementary Fig. 49). 500 standard events of sucrose were collected as a training dataset and fed to a One-Class SVM algorithm. This trained algorithm was further applied in the analysis of the outlier events acquired with the Green kiwifruit sample, by which few sucrose events were recognized according to the consistency of event features (Supplementary Fig. 50). The outlier analysis for sucrose detected in the Green kiwifruit was also demonstrated in Supplementary Fig. 51. By doing that, we can confirm that sucrose was detected in both Green and Sungold kiwifruit. The events of sucrose are more clearly seen in the Sungold kiwifruit likely because a higher concentration of sucrose is present in Sungold kiwifruit. However, events of sucrose were not detected in Rubyred kiwifruit, even assisted by searching using machine learning. However, two extra clusters of events were detected in Rubyred kiwifruit. Though these two event types were inconsistent with any previously identified cis-diols using MspA-PBA, they are treated as characteristic event types specifically detected in this kiwifruit variety. For the ease of demonstration, they were respectively marked as U1 (unidentified cis-diol 1) and U2 (unidentified cis-diol 2).

To summarize, cis-diols identified in Green kiwifruit and Sungold kiwifruit were confirmed to be sucrose, D-FRU, D-GLC, L-MA, L-TA, CA and ICIT (Fig. 6b, c, e, f). The results are also consistent with that previously studied using conventional methods^61,63. For Rubyred kiwifruit, events of D-FRU, D-GLC, L-MA, L-TA, ICIT, U1 and U2 were identified (Fig. 6h, i). Three independent trails were also carried out with each kiwifruit variety to show the consistency of measurement (Supplementary Figs 52–54). According to above results, events of D-FRU, D-GLC, L-MA, L-TA and ICIT were detected from all kiwifruit varieties. Sucrose was only detected in Green and Sungold kiwifruit and the Sungold kiwifruit reports a higher concentration of sucrose than that measured with the Green kiwifruit and sucrose was not detected in Rubyred kiwifruit. Instead, two previously unidentified cis-diols were detected only in Rubyred kiwifruit. These different combinations of cis-diols in different kiwifruit varieties should contribute to the different taste of these kiwifruit products. However, to the best of our knowledge, a direct single molecule discrimination between molecular compounds in different fruit varieties has never been previously reported.

Rapid analysis of commercial fruit juice products

Following the same principle, cis-diols may as well be identified directly from commercial fruit juice products. Specifically, L-malic acid (L-MA), which is a naturally occurring α-hydroxy acid, is widely found in fruits⁶⁴. Its enantiomer, D-malic acid (D-MA), is however rarely seen in nature. Due to the cost reasons, a mixture of DL-malic acid (Fig. 7a) instead of L-malic acid, is more frequently applied as synthetic food additive^65,66. Another pair of similar α-hydroxy acid enantiomers used as food additive is DL-tartaric acid⁶⁷. However, rapid discrimination of α-hydroxy acid enantiomers directly from a food sample is not a trivial task.

**Fig. 7: Rapid analysis of commercial juice and juice drink.**

To testify this using nanopore, MspA-PBA sensing of D-MA and L-MA was respectively carried out in a 1.5 M KCl buffer with a + 160 mV bias. Representative traces were demonstrated in Fig. 7b, according to which the events of D-MA and L-MA are fully distinguishable. This difference is more clearly shown in the corresponding event scatter plot (Fig. 7c, Supplementary Figs 8, 55), further confirming the high resolution of MspA-PBA in the discrimination between α-hydroxy acid enantiomers. To enable machine learning based identification, 500 events acquired with D-MA were collected and added to the existing training set already containing ten cis-diols (Supplementary Fig. 56). By doing that, a new machine learning model, which can identify 11 cis-diols including both D-MA and L-MA, is established. According to the training and validation with a variety of models, the Bagging Trees model has performed the best by reporting an accuracy of 98.9% (Supplementary Fig. 57). This model is thus used for all further event predictions.

Nanopore sensing was then respectively carried out with commercial 100% grape juice (Huiyuan^TM, Fig. 7d) and grape juice drink (Wahaha^TM, Fig. 7g). The measurements were performed with MspA-PBA in a 1.5 M KCl buffer and a + 160 mV continually applied bias. Without performing any sample separation, 2 μL different types of juice sample was respectively added to the nanopore device in separate measurements. Immediately afterwards, corresponding nanopore events were observed with different commercial juice products (Supplementary Figs 58, 59). Three independent trails were also carried out for each condition to show its reproducibility (Supplementary Figs 60–61).

All acquired nanopore events were first treated by outlier analysis (Supplementary Figs 62, 63). By using One-Class SVM, an unsupervised machine learning algorithm for outlier detection, all events that fail to show any consistency with the standard events generated by the previously investigated 11 cis-diols, were grouped as outlier events. All outlier events were further treated by cluster analysis to check the presence of any cluster-forming events. Specifically, for the events acquired with the 100% grape juice, a cluster of events which has never been previously investigated using MspA-PBA, was discovered. For the ease of demonstration, this cluster of events is referred to as U3 (unidentified cis-diol 3). The U3 events were however never detected at all in any freshly squeezed grape juice samples mentioned in this paper. It is speculated that the U3 event might be from the specific variety of grape used in the generation of the commercial 100% grape juice or it might be from food additives introduced during juice production.

For all inlier events, the newly established Bagging Trees model was applied for event identification. For both types of commercial juice products, events of D-FRU, D-GLC, L-MA and L-TA were clearly identified (Fig. 7e–f, h–i). Three independent trails were also carried out with each commercial juice product to show the consistency of measurement (Supplementary Figs 64–65). In the 100% grape juice, only L-MA was detected, suggesting that no DL-MA food additive was added to this juice product. However, both D-MA and L-MA were detected in the grape juice drink, consistent with that listed in the corresponding ingredient table. With the established machine learning algorithm, all above fruit juice analysis was carried out extremely rapidly. Both juice samples were added directly to the nanopore device without any need of sample separation.

To also show the portability of the technique, nanopore analysis of fruits was also carried out with an Orbit mini portable patch clamp amplifier (Nanion Technologies GmbH) (Fig. 8). To perform the measurement, the Orbit mini, which is of a palm size and an extremely light weight, is connected to a routine laptop computer, weighing ~1 kg. The operation of the whole setup can be powered solely by the inbuilt laptop battery, enabling the application of nanopore sensing even without any electricity supply. Representative traces and statistical results produced with natural grape juice at this condition are summarized in Fig. 8. The results are generally consistent with that acquired using the Axon200B patch clamp amplifier and Digidata 1550B digitizer.

**Fig. 8: Portable nanopore analysis of grape juice.**

Compared with conventional analytical methods such as mass spectrometry, liquid chromatography and gas chromatography, a nanopore platform generally offers a higher portability, a lower measurement cost, a more simplified sample preparation and the capacity to perform simultaneous sensing of multiple analytes, which is extremely suitable for portable food analysis. This might be useful for food industry or food and drug administration. Specifically, its high resolution in direct discrimination between isomers and enantiomers is advantageous over MS. It also offers a more rapid result feedback, which is generally faster than chromatography-based methods. Being an emerging method, the nanopore platform at its present form is still disadvantageous in the limit of detection and the accuracy of quantification. We however foresee that these disadvantages could be improved in future development of this technique.

Discussion

MspA-PBA, a reactive MspA nanopore containing a sole PBA adapter, was applied to the identification in fruit juice of catechin, neochlorogenic acid, D-sorbitol, xylitol, DL-malic acid, L-tartaric acid, citric acid, (2R,3S)-isocitric acid, sucrose, D-glucose and D-fructose. These 12 analytes respectively represent saccharides, alditols, 1,2-diphenols and α-hydroxy acids widely encountered in fruits. Though nucleotides are in principle detectable by MspA-PBA^42,44, their concentrations are too low to be detected directly from the fruit juice. In each measurement, only ~ µL fruit juice was added, which doesn’t significantly change the pH and ion strength of the buffer environment (Supplementary Table 6). The inherent proteins, nucleic acids and polysaccharides in fruits, which either don’t interact with the PBA or report clearly distinguishable events, are also not interfering with the measurements. By simultaneously considering five event features, including ΔI/I₀, std, kurt, skew and time, a custom machine learning algorithm provides automatic and unbiased identification of unknown nanopore events with a classification accuracy of 99.3%. A nanopore that can simultaneously detect saccharides, alditols, 1,2-diphenols and α-hydroxy acids has not been demonstrated previously. Although the previous studies have reported the use of phenylboronic acid-modified nanopores for saccharide sensing, they either fail to achieve a single molecule resolution and are incapable to perform simultaneous sensing^68,69 or report a clearly insufficient resolution⁷⁰ for fruit analysis. This sensing capacity of MspA-PBA was further applied to the direct and rapid analysis of natural fruit juice of prune, grape, lemon and different varieties of kiwifruits. The capacity of MspA-PBA to discriminate between DL-malic acid also enables its immediate application in the detection of synthetic food additive in commercial juice products. To the best of our knowledge, nanopore sensing of catechin, neochlorogenic acid, DL-malic acid, L-tartaric acid, citric acid, (2R,3S)-isocitric acid and sucrose has not been previously reported. Nanopore analysis of natural fruit juice, which systematically demonstrates how natural complex samples can be analyzed in single molecule, was also not previously reported. Though the technique at its current form can only detect free cis-diols in natural samples, it is possible to release the bound cis-diol by sample treatment of enzymatic^71,72 or chemical hydrolysis^71,73,74, followed with nanopore analysis. Though only demonstrated with fruits, this sensing strategy is in principle generally suitable for other natural samples containing cis-diol components. All nanopore measurements can also be carried out in a miniaturized device (Fig. 8) to demonstrate portable food analysis. Different from that required in genome sequencing, the food samples normally contain abundant cis-diols and the need of a high-throughput acquisition system to maximize the efficiency of sensing is less urgent. However, a multiplexed array of sensors which can simultaneously analyze different food samples is more practically needed.

Methods

Nanopore preparations

The preparation of MspA-PBA^36,42,43,44 was carried out as described below. Briefly, the desired MspA hetero-octamer, which was referred to as (N90C)₁(M2)₇, was composed of seven units of M2 MspA-D16H6 (D90N/D91N/D93N/D118R/D134R/E139K) and one unit of N90C MspA-H6 (D90C/D91N/D93N/D118R/D134R/E139K). The genes of M2 MspA-D16H6 and N90C MspA-H6 were respectively synthesized by GenScript. Both genes were simultaneously placed in the same co-expression vector pETDuet-1 and expressed in E. coli BL21 (DE3) pLysS competent cells (GenScript, New Jersey). After protein expression, the collected bacterial pellets were resuspended in the lysis buffer (100 mM Na₂HPO₄/NaH₂PO₄, 0.1 mM EDTA, 150 mM NaCl, 0.5% (v/v) Genapol X-80, pH 6.5) and incubated at 60 °C for 50 min. After being centrifuged and ice-incubated for 30 min, the supernatant in the suspension was collected, filtered and loaded to a HisTrap^TMHP nickel ion affinity column (GE Healthcare). The column was first eluted with buffer A (0.5 M NaCl, 20 mM HEPES, 5 mM imidazole, 0.5% (w/v) Genapol X-80, pH 8.0) to elute miscellaneous proteins. It was then eluted with a linear gradient of mixing buffer A and buffer B (0.5 M NaCl, 20 mM HEPES, 500 mM imidazole, 0.5% (w/v) Genapol X-80, pH 8.0) to collect all hetero-octameric assemblies of MspAs. To separate (N90C)₁(M2)₇ from all other hetero-octameric assemblies, a 10% SDS- polyacrylamide gel was used to perform gel electrophoresis of the eluent fractions collected from the nickel column. After gel electrophoresis, the gel was stained with coomassie brilliant blue (1.25 g coomassie brilliant blue R250, 225 mL methanol, 50 mL glacial acetic acid, 225 mL ultrapure water) for 4 h and de-stained with an elution buffer (40% (v/v) MeOH, 10% (v/v) glacial AcOH). The gel fragment containing the target (N90C)₁(M2)₇ assembly was then excised from the gel. The obtained gel piece was crushed and immersed in a protein extraction buffer (150 mM NaCl, 15 mM Tris-HCl, 0.2% (w/v) DDM, 0.5% (v/v) Genapol X-80, 5 mM TCEP, 10 mM EDTA, pH 7.5). The extracted (N90C)₁(M2)₇ was immediately used or stored at −80 °C for long term use.

To modify (N90C)₁(M2)₇ with phenylboronic acid, 2 μL 3-(maleimide) phenylboronic acid (MPBA, 500 mM) and 5 μL freshly prepared (N90C)₁(M2)₇ were simultaneously added to a 43 μL 1.5 M KCl buffer (1.5 M KCl, 100 mM MOPS, pH 7.0) and thoroughly mixed. The mixture was incubated at 25 °C for 10 min to finalize pore modification. The prepared nanopore was either immediately used in all downstream nanopore measurements or is stored at −80 °C for long term use. For simplicity, this MPBA modified (N90C)₁(M2)₇ is referred to as MspA-PBA throughout the manuscript.

Nanopore measurements

Nanopore measurements were performed similarly to that reported previously⁴². All nanopore measurements were performed with an Axonpatch 200B patch-clamp amplifier paired with a Digidata 1550B digitizer (Molecular Devices). To reduce external environment noises, the custom measurement device was fixed in a custom Faraday cage mounted on an optical table (Jiangxi Liansheng technology Co., Ltd). The measurement device was composed of two chambers separated by a Teflon film containing an aperture of about 100 μm in diameter. Prior to the measurement, the aperture was first treated with 2% (v/v) hexadecane in pentane and set for pentane evaporation. Then both chambers were filled with 500 μL 1.5 M KCl buffer (1.5 M KCl, 100 mM MOPS, pH 7.0). Afterwards, two Ag/AgCl electrodes which were separately connected to the patch-clamp amplifier, were respectively inserted in each chamber, in contact with the buffer solution. By convention, the electrically grounded chamber is defined as the cis chamber and its opposing chamber is defined as the trans chamber. To form a phospholipid bilayer on the aperture of the Teflon film, 100 µL pentane solution of DPhPC (5 mg/mL) was added to both chambers and the lipid bilayer spontaneously forms when the buffer solution in one of the chambers has been pipetted up and down for several times. Spontaneous pore insertion was triggered by adding newly prepared MspA-PBA to the cis chamber. Upon a single nanopore insertion, the buffer in the cis chamber was immediately exchanged with fresh buffer to avoid further pore insertions.

All nanopore measurements were performed with a single MspA-PBA channel and all analytes were added to both chambers. Specifically, the portable nanopore analysis of natural grape juice analysis was carried out using Nanion Orbit Mini (Fig. 8). Unless otherwise stated, all other measurements were performed using Axonpatch 200B patch-clamp amplifier paired with a Digidata 1550B digitizer (Molecular Devices). Unless otherwise stated, all measurements were performed at 25 degrees of Celsius with a continually applied bias of +160 mV. All single-channel recordings were sampled at 25 kHz and low-pass filtered with a corner frequency of 1 kHz. Unless otherwise stated, the final concentrations of the analytes in both chambers were set as: 0.8 mM for catechin (CAT), 0.5 mM for neochlorogenic acid (3-CQA), 2 mM for D-sorbitol (D-SOR), 2 mM for xylitol (XYL), 0.6 mM for L-malic acid (L-MA) and D-malic acid (D-MA), 0.6 mM for L-tartaric acid (L-TA), 6 mM for citric acid (CA), 1 mM for isocitric acid (ICIT), 30 mM for D-glucose (D-GLC) and 10 mM for D-fructose (D-FRU). Based on the principle of nanopore sensing by single molecule reaction, the event feature is independent of the analyte concentration. With only a sole reactive adapter at the pore constriction, simultaneous capturing of more than one analyte is also technically impossible. However, some of the analyte type may require a higher analyte concentration to report a sufficiently high event appearance rate. Thus, the selection of these analyte concentrations is to maximize the efficiency of nanopore appearance rate.

Data analysis

Event detection was first performed by the “single channel search” function of Clampfit 10.7. Subsequently, the start time, end time and dwell time of each event were obtained. Only events with a dwell time more than 10 ms were used for further analysis. The start and end times of events were listed in a text file for subsequent event feature extraction. The raw Axon abf files and the text file containing the start and end times of the events in the files were imported into MATLAB. Based on the start time and the end time of the event, the event features including blockage ratio (ΔI/I₀), standard deviation (std), kurtosis (kurt), skewness (skew) and dwell time (time) were extracted for machine learning. The generation of histograms and scatter plots were performed by Origin 9.2 (Origin Lab).

Machine learning was performed by the “classification learner” toolbox in MATLAB. For CAT, 3-CQA, D-SOR, XYL, L-MA, L-TA, ICIT, 500 and 100 events of each analyte were collected to form the training set and the testing set, respectively. For D-GLC and D-FRU, 1000 and 200 events of each analyte were collected to form the training and the testing set. For CA, 100 and 20 events were collected to form the training and the testing set. The label of each event was assigned as the identification of the chemical compound used in the production of the data. Model training was performed with the training set and the 10-fold cross validation accuracies were obtained. The testing set was used to test all trained models and testing accuracies were reported. Different models, including Decision Trees, Discriminant Analysis, Naïve Bayes, Support Vector Machine (SVM), K Nearest Neighbor (KNN), Ensemble and Neural Network were trained and tested. According to the results of 10-fold cross validation accuracy and the testing accuracy (Supplementary Table 3), the Bagging Trees model reported the highest overall accuracy. Therefore, the trained Bagging Trees model was used for further event prediction.

The density-based spatial clustering of applications with noise (DBSCAN) cluster analysis was applied for clustering on a Python platform. The Epsilon was set to 0.17 and the min_samples was set to 18 for analysis of prune juice, grape juice and lemon juice. The Epsilon was set to 0.1 and the min_samples was set to 10 for analysis of kiwifruit. The Epsilon was set to 0.3 and the min_samples was set to 30 for analysis of commercial 100% grape juice.

One-Class SVM, a machine learning algorithm for outlier analysis, was performed by the ‘Scikit-learn library’ in Python. The training set for outlier analysis of kiwifruits is the same as the ten cis-diols training set described in Fig. 3. The training set for outlier analysis of commercial products is the same as the eleven cis-diols training set described in Supplementary Fig. 57. The training set of sucrose, which contains 500 standard events was acquired with sucrose. The data was later used in the identification of sucrose in Sungold kiwifruit and Green kiwifruit. Briefly, ΔI/I₀, std, kurt, skew and time of events were employed as event features for outlier analysis. The parameter ‘nu’ was set to 0.05. Different One-Class SVM models were respectively trained by different training set for further event discrimination. The model automatically judges whether the analyzed event belongs to any previously trained event types so that the event is considered as an inlier event. If not, it is considered as an outlier event instead. The inlier events of kiwifruits and commercial products were further predicted by the trained Bagging Trees models, as respectively described in Fig. 3 and Supplementary Fig. 57. The outlier events were then treated with cluster analysis using DBSCAN. If a group of these outlier events were identified to have similar event features and appear as a cluster, this group of outlier events are considered a set of detected but unidentified cis-diol.

All machine learning models and sample data used to train, validate and test the model have been deposited in Figshare. Please follow the link https://figshare.com/s/5fc838dad55e6452bc95 for download.

Fruit juice preparations

Technical details of all fruit samples used in the measurements were demonstrated in Supplementary Fig. 67. To perform the measurement, different fruits were respectively washed, cut into pieces and mashed into pulp. Then, the pulp of different fruits was separately treated by a mini food processor to produce puree (Figs 4a, 5a, 6). To remove the solid residues, the puree was centrifuged at 1900 g for 10 min at 4 °C and the supernatant was collected as juice. The supernatant was then added into an ultrafiltration tube with a 3 kDa molecular weight cut off (MWCO) and centrifuged at 2350 g for 10 min at 4 °C. The filtrate was collected and immediately used in all downstream measurements. The whole operation takes about 25 min (Fig. 4a).

In Supplementary Fig. 37, the lemons were cut into fruit pieces. The fruit pieces were pinched by hand to product fruit juice. And the fruit juice was added into the ultrafiltration tube with a 3 kDa molecular weight cut off (MWCO) and centrifuged at 3380 g for 3 min at 4 °C. The filtrate was collected and immediately used in all downstream measurements. The whole operation takes about 8 min (Supplementary Fig. 37a).

Quantification of target cis-diols

The concentrations of target cis-diols were evaluated according to the following equation:

$${C}_{i}={E}_{i}/\left({k}_{i}*t\right)$$

(1)

Where $i$ (from 1 to 10) stands for different cis-diols, including CAT, 3-CQA, D-SOR, XYL, L-MA, L-TA, CA, ICIT, D-GLC and D-FRU, respectively. ${E}_{i}$ is the number of binding events of each cis-diol type. ${k}_{i}$ is the calibration coefficient defined as the number of event occurrences per unit concentration (mM) per unit time (ms). The value of ${k}_{i}$ is derived from results of the concentration dependence measurements (Supplementary Figs 14–23). The results were also summarized in Supplementary Tables 4, 5. $t$ is the recording time of the measurement.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Data supporting the findings of this study are shown in the main text and the Supplementary Information, which are also available within the source data provided with this paper. All data used to train, evaluate and test the machine learning model have been deposited in Figshare. Please follow the link: https://figshare.com/s/5fc838dad55e6452bc95 for download. Source data are provided with this paper.

Code availability

The custom machine learning code and sample data used to train, validate and test the model have been deposited in Figshare. For simplicity, the shared package is referred to as ‘fruit classifier’. Please follow the link https://figshare.com/s/5fc838dad55e6452bc95 for download.

References

Issaoui, M. et al. Phenols, flavors, and the mediterranean diet. J. Aoac. Int. 103, 915–924 (2020).
Article PubMed Google Scholar
Al-Farsi, M. et al. Quality evaluation of Omani honey. Food Chem. 262, 162–167 (2018).
Article CAS PubMed Google Scholar
Fang, T. et al. Analysis of sorbitol content variation in wild and cultivated apples. J. Sci. Food Agric. 100, 139–144 (2020).
Article CAS PubMed Google Scholar
Tiitinen, K. M., Yang, B., Haraldsson, G. G., Jonsdottir, S. & Kallio, H. P. Fast analysis of sugars, fruit acids, and vitamin C in sea buckthorn (Hippophae rhamnoides L.) varieties. J. Agric. Food Chem. 54, 2508–2513 (2006).
Article CAS PubMed Google Scholar
Malongane, F., McGaw, L. J. & Mudau, F. N. The synergistic potential of various teas, herbs and therapeutic drugs in health improvement: a review. J. Sci. Food Agric. 97, 4679–4689 (2017).
Article CAS PubMed Google Scholar
Erejuwa, O. O., Sulaiman, S. A. & Wahab, M. S. Fructose might contribute to the hypoglycemic effect of honey. Molecules 17, 1900–1915 (2012).
Article CAS PubMed PubMed Central Google Scholar
Scholz, D., Brooks, G. J., Parish, D. F. & Burmeister, F. Fruit acid extracts, a fresh approach to skin renewal. Int. J. Cosmet. Sci. 16, 265–272 (1994).
Article CAS PubMed Google Scholar
Oroian, M., Ropciuc, S. & Paduret, S. Honey adulteration detection using raman spectroscopy. Food Anal. Methods 11, 959–968 (2017).
Article Google Scholar
Iriondo-DeHond, A. et al. Validation of coffee by-products as novel food ingredients. Innov. Food Sci. Emerg. 51, 194–204 (2019).
Article CAS Google Scholar
Akagic, A. et al. Geographic variability of sugars and organic acids in selected wild fruit species. Foods 9, 462 (2020).
Palma-Duran, S. A. et al. A Comprehensive HPLC-DAD-ESI-MS Validated Method for the Quantification of 16 Phytoestrogens in Food, Serum and Urine. Appl. Sci. 10, 8147 (2020).
Fernandez, P. L., Martin, M. J., Gonzalez, A. G. & Pablos, F. HPLC determination of catechins and caffeine in tea. Differentiation of green, black and instant teas. Analyst 125, 421–425 (2000).
Article ADS CAS PubMed Google Scholar
Ouchemoukh, S., Schweitzer, P., Bachir Bey, M., Djoudad-Kadji, H. & Louaileche, H. HPLC sugar profiles of Algerian honeys. Food Chem. 121, 561–568 (2010).
Article CAS Google Scholar
Muir, J. G. et al. Measurement of short-chain carbohydrates in common Australian vegetables and fruits by high-performance liquid chromatography (HPLC). J. Agric. Food Chem. 57, 554–565 (2009).
Article CAS PubMed Google Scholar
Mikulic-Petkovsek, M., Koron, D. & Rusjan, D. The impact of food processing on the phenolic content in products made from juneberry (Amelanchier lamarckii) fruits. J. Food Sci. 85, 386–393 (2020).
Article CAS PubMed Google Scholar
Carneiro Ferreira, I. et al. Brazilian varieties of acerola (Malpighia emarginata DC.) produced under tropical semi-arid conditions: Bioactive phenolic compounds, sugars, organic acids, and antioxidant capacity. J. Food Biochem. 45, e13829 (2021).
Article CAS Google Scholar
Coyago-Cruz, E. et al. Effect of the fruit position on the cluster on fruit quality, carotenoids, phenolics and sugars in cherry tomatoes (Solanum lycopersicum L.). Food Res. Int. 100, 804–813 (2017).
Article CAS PubMed Google Scholar
Jiang, C. C., Fang, Z. Z., Zhou, D. R., Pan, S. L. & Ye, X. F. Changes in secondary metabolites, organic acids and soluble sugars during the development of plum fruit cv. ‘Furongli’ (Prunus salicina Lindl). J. Sci. Food Agric. 99, 1010–1019 (2019).
Article CAS PubMed Google Scholar
Nowicka, P., Wojdylo, A. & Teleszko, M. Effect of mixing different kinds of fruit juice with sour cherry puree on nutritional properties. J. Food Sci. Technol. 54, 114–129 (2017).
Article CAS PubMed PubMed Central Google Scholar
Fernández-Fernández, R. et al. Simple LC–MS determination of citric and malic acids in fruits and vegetables. Chromatographia 72, 55–62 (2010).
Article Google Scholar
Zhou, Y. et al. Fruit sugar and organic acid were significantly related to fruit Mg of six citrus cultivars. Food Chem. 259, 278–285 (2018).
Article CAS PubMed Google Scholar
Morvai, M. & Molnár-Perl, I. Simultaneous gas chromatographic quantitation of sugars and acids in citrus fruits, pears, bananas, grapes, apples and tomatoes. Chromatographia 34, 502–504 (1992).
Article CAS Google Scholar
Manrao, E. A. et al. Reading DNA at single-nucleotide resolution with a mutant MspA nanopore and phi29 DNA polymerase. Nat. Biotechnol. 30, 349–353 (2012).
Article CAS PubMed PubMed Central Google Scholar
Li, Q., Ying, Y. L., Liu, S. C., Lin, Y. & Long, Y. T. Detection of single proteins with a general nanopore sensor. ACS Sens 4, 1185–1189 (2019).
Article CAS PubMed Google Scholar
Si, W. & Aksimentiev, A. Nanopore sensing of protein folding. ACS Nano 11, 7091–7100 (2017).
Article CAS PubMed PubMed Central Google Scholar
Liu, Y. et al. Machine learning assisted simultaneous structural profiling of differently charged proteins in a mycobacterium smegmatis porin A (MspA) electroosmotic trap. J. Am. Chem. Soc. 144, 757–768 (2022).
Article CAS PubMed Google Scholar
Wang, Y. et al. Structural-profiling of low molecular weight RNAs by nanopore trapping/translocation using Mycobacterium smegmatis porin A. Nat. Commun. 12, 3368 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Liao, D. F., Cao, C., Ying, Y. L. & Long, Y. T. A general strategy of aerolysin nanopore detection for oligonucleotides with the secondary structure. Small 14, e1704520 (2018).
Article PubMed Google Scholar
Lucas, F. L. R. et al. The manipulation of the internal hydrophobicity of FraC nanopores augments peptide capture and recognition. ACS Nano 15, 9600–9613 (2021).
Article CAS PubMed PubMed Central Google Scholar
Piguet, F. et al. Identification of single amino acid differences in uniformly charged homopolymeric peptides with aerolysin nanopore. Nat. Commun. 9, 966 (2018).
Article ADS PubMed PubMed Central Google Scholar
Versloot, R. C. A., Straathof, S. A. P., Stouwie, G., Tadema, M. J. & Maglia, G. beta-Barrel nanopores with an acidic-aromatic sensing region identify proteinogenic peptides at low pH. ACS Nano 16, 7258–7268 (2022).
Article CAS PubMed PubMed Central Google Scholar
Cao, J. et al. Giant single molecule chemistry events observed from a tetrachloroaurate(III) embedded Mycobacterium smegmatis porin A nanopore. Nat. Commun. 10, 5668 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Boersma, A. J., Brain, K. L. & Bayley, H. Real-time stochastic detection of multiple neurotransmitters with a protein nanopore. ACS Nano 6, 5304–5308 (2012).
Article CAS PubMed Google Scholar
Kasianowicz, J., Walker, B., Krishnasastry, M. & Bayley, H. Genetically-engineered pores as metal-ion biosensors. Mater. Res. Soc. Symp. Proc. 330, 217–223 (1994).
Article CAS Google Scholar
Braha, O. et al. Designed protein pores as components for biosensors. Chem. Biol. 4, 497–505 (1997).
Article CAS PubMed Google Scholar
Zhang, S. et al. A nanopore-based saccharide sensor. Angew. Chem. Int. Ed. 61, e202203769 (2022).
Article ADS CAS Google Scholar
Wang, S. et al. Single molecule observation of hard-soft-acid-base (HSAB) interaction in engineered Mycobacterium smegmatis porin A (MspA) nanopores. Chem. Sci. 11, 879–887 (2019).
Article PubMed PubMed Central Google Scholar
Cao, J. et al. A single-molecule observation of dichloroaurate(I) binding to an engineered mycobacterium smegmatis porin A (MspA) nanopore. Anal. Chem. 93, 1529–1536 (2021).
Article CAS PubMed Google Scholar
Liaudanskas, M. et al. Variability in the content of phenolic compounds in plum fruit. Plants (Basel) 9, 1611 (2020).
Article CAS PubMed Google Scholar
Filip, M., Vlassa, M., Coman, V. & Halmagyi, A. Simultaneous determination of glucose, fructose, sucrose and sorbitol in the leaf and fruit peel of different apple cultivars by the HPLC-RI optimized method. Food Chem. 199, 653–659 (2016).
Article CAS PubMed Google Scholar
Goudie, A. J. & Rieman, W. Chromatographic separation and determination of fruit acids. Anal. Chim. Acta 26, 419–423 (1962).
Article CAS Google Scholar
Wang, Y. et al. Identification of nucleoside monophosphates and their epigenetic modifications using an engineered nanopore. Nat. Nanotechnol. 17, 976–983 (2022).
Article ADS CAS PubMed Google Scholar
Liu, Y. et al. Nanopore identification of alditol epimers and their application in rapid analysis of alditol-containing drinks and healthcare products. J. Am. Chem. Soc. 144, 13717–13728 (2022).
Article CAS PubMed Google Scholar
Wang, Y. et al. Discrimination of ribonucleoside Mono-, Di-, and triphosphates using an engineered nanopore. ACS Nano 16, 21356–21365 (2022).
Article CAS PubMed Google Scholar
Pappin, B., J, M. & A, T. Carbohydrates - Comprehensive Studies on Glycobiology and Glycotechnology, Ch. 3 (IntechOpen, 2012).
Jia, W. et al. Programmable nano-reactors for stochastic sensing. Nat. Commun. 12, 5811 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
van den Berg, R., Peters, J. A. & van Bekkum, H. The structure and (local) stability constants of borate esters of mono- and di-saccharides as studied by 11B and 13C NMR spectroscopy. Carbohydr. Res. 253, 1–12 (1994).
Article PubMed Google Scholar
Butler, T. Z., Pavlenok, M., Derrington, I. M., Niederweis, M. & Gundlach, J. H. Single-molecule DNA detection with an engineered MspA protein nanopore. Proc. Natl Acad. Sci. USA 105, 20647–20652 (2008).
Article ADS CAS PubMed PubMed Central Google Scholar
Niederweis, M. et al. Cloning of the mspA gene encoding a porin from Mycobacterium smegmatis. Mol. Microbiol. 33, 933–945 (1999).
Article CAS PubMed Google Scholar
Hahsler, M., Piekenbrock, M. & Doran, D. dbscan: fast density-based clustering with R. J. Stat. Softw. 91, 1–30 (2019).
Article Google Scholar
Li, J., Zhang, C., Liu, H., Liu, J. & Jiao, Z. Profiles of sugar and organic acid of fruit juices: a comparative study and implication for authentication. J. Food Qual. 2020, 1–11 (2020).
Google Scholar
Chusid, M. J. & Chusid, J. A. Diarrhea induced by sorbitol. J. Pediatr. 99, 326 (1981).
Article CAS PubMed Google Scholar
Park, S. Y., Jin, M. L., Yi, E. H., Kim, Y. & Park, G. Neochlorogenic acid inhibits against LPS-activated inflammatory responses through up-regulation of Nrf2/HO-1 and involving AMPK pathway. Environ. Toxicol. Pharmacol. 62, 1–10 (2018).
Article CAS PubMed Google Scholar
Zhao, Q. et al. Investigating the role of tartaric acid in wine astringency. Food Chem. 403, 134385 (2023).
Article CAS PubMed Google Scholar
Wang, H. et al. Effect of sodium sulfite, tartaric acid, tannin, and glucose on rheological properties, release of aroma compounds, and color characteristics of red wine. Food Sci. Biotechnol. 28, 395–403 (2019).
Article CAS PubMed Google Scholar
Veldhuizen, M. G., Siddique, A., Rosenthal, S. & Marks, L. E. Interactions of lemon, sucrose and citric acid in enhancing citrus, sweet and sour flavors. Chem. Senses 43, 17–26 (2017).
Article PubMed PubMed Central Google Scholar
Sato, M. et al. Characterization of five Indonesian mangoes using gas chromatography-mass spectrometry-based metabolic profiling and sensory evaluation. J. Biosci. Bioeng. 132, 613–620 (2021).
Article CAS PubMed Google Scholar
Yan, S. et al. Rapid and multiplex preparation of engineered Mycobacterium smegmatis porin A (MspA) nanopores for single molecule sensing and sequencing. Chem. Sci. 12, 9339–9346 (2021).
Article CAS PubMed PubMed Central Google Scholar
Heinz, C., Engelhardt, H. & Niederweis, M. The core of the tetrameric mycobacterial porin MspA is an extremely stable β-Sheet domain*. J. Biol. Chem. 278, 8678–8685 (2003).
Article CAS PubMed Google Scholar
Heinz, C. & Niederweis, M. Selective extraction and purification of a mycobacterial outer membrane protein. Anal. Biochem. 285, 113–120 (2000).
Article CAS PubMed Google Scholar
Sivakumaran, S., Huffman, L., Sivakumaran, S. & Drummond, L. The nutritional composition of Zespri(R) SunGold Kiwifruit and Zespri(R) sweet green kiwifruit. Food Chem. 238, 195–202 (2018).
Article CAS PubMed Google Scholar
Yuan, X. et al. Comparative study on physicochemical and nutritional qualities of kiwifruit varieties. Foods 12, 108 (2023).
Article CAS Google Scholar
Henare, S. J. Nutritional Composition of Fruit Cultivars, Ch. 15, 337–370 (Academic Press, 2016).
Wei, Z. et al. Microbial biosynthesis of L-Malic acid and related metabolic engineering strategies: advances and prospects. Front. Bioeng. Biotechnol. 9, 765685 (2021).
Article PubMed PubMed Central Google Scholar
Laganà, P. et al. Use and Overuse of Food Additives in Edible Products: Health Consequences for Consumers, Chemistry and Hygiene of Food Additives, 39–46 (Springer Cham, 2017).
Wu, L., Zhang, Q., Shan, L. & Chen, Z. Identifying critical factors influencing the use of additives by food enterprises in China. Food Control 31, 425–432 (2013).
Article Google Scholar
Kodama, S., Yamamoto, A., Matsunaga, A. & Hayakawa, K. Direct chiral resolution of tartaric acid in food products by ligand exchange capillary electrophoresis using copper(II)-D-quinic acid as a chiral selector. J. Chromatogr. A. 932, 139–143 (2001).
Article CAS PubMed Google Scholar
Yang, M. et al. Rational design of stimuli-responsive polymers modified nanopores for selective and sensitive determination of salivary glucose. Anal. Chem. 91, 14029–14035 (2019).
Article CAS PubMed Google Scholar
Nguyen, Q. H., Ali, M., Neumann, R. & Ensinger, W. Saccharide/glycoprotein recognition inside synthetic ion channels modified with boronic acid. Sens. Actuators, B 162, 216–222 (2012).
Article CAS Google Scholar
Ramsay, W. J. & Bayley, H. Single-molecule determination of the isomers of d-glucose and d-fructose that bind to boronic acids. Angew. Chem. Int. Ed. 57, 2841–2845 (2018).
Article CAS Google Scholar
Ruhaak, L. R. et al. Glycan labeling strategies and their use in identification and quantification. Anal. Bioanal. Chem. 397, 3457–3481 (2010).
Article CAS PubMed PubMed Central Google Scholar
Robbins, R. J. Phenolic acids in foods:an overview of analytical methodology. J. Agric. Food Chem. 51, 2866–2887 (2003).
Article CAS PubMed Google Scholar
Mattila, P. & Kumpulainen, J. Determination of free and total phenolic acids in plant-derived foods by HPLC with diode-array detection. J. Agric. Food Chem. 50, 3660–3667 (2002).
Article CAS PubMed Google Scholar
Yu, J. et al. Phenolic profiles, bioaccessibility and antioxidant activity of plum (Prunus Salicina Lindl). Food Res. Int. 143, 110300 (2021).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

This project was funded by National Natural Science Foundation of China (Grant No. 22225405, No. 31972917, to S.H.), the National Key R&D Program of China (Grant No. 2022YFA1304602, to S.H.), the Fundamental Research Funds for the Central Universities (Grant No.020514380257, to S.H.), Programs for high-level entrepreneurial and innovative talents introduction of Jiangsu Province (individual and group program, to S.H.), Natural Science Foundation of Jiangsu Province (Grant No. BK20200009, to S.H.), Excellent Research Program of Nanjing University (Grant No. ZYJH004, to S.H.). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

Author information

These authors contributed equally: Pingping Fan, Zhenyuan Cao, Shanyu Zhang.

Authors and Affiliations

State Key Laboratory of Analytical Chemistry for Life Sciences, School of Chemistry and Chemical Engineering, Nanjing University, 210023, Nanjing, China
Pingping Fan, Zhenyuan Cao, Shanyu Zhang, Yuqin Wang, Yunqi Xiao, Wendong Jia, Panke Zhang & Shuo Huang
Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, 210023, Nanjing, China
Pingping Fan, Zhenyuan Cao, Shanyu Zhang, Yuqin Wang, Yunqi Xiao, Wendong Jia & Shuo Huang
State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, 210023, Nanjing, China
Yuqin Wang
Institute for the Environment and Health, Nanjing University Suzhou Campus, 215163, Suzhou, China
Yuqin Wang

Authors

Pingping Fan
View author publications
You can also search for this author in PubMed Google Scholar
Zhenyuan Cao
View author publications
You can also search for this author in PubMed Google Scholar
Shanyu Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Yuqin Wang
View author publications
You can also search for this author in PubMed Google Scholar
Yunqi Xiao
View author publications
You can also search for this author in PubMed Google Scholar
Wendong Jia
View author publications
You can also search for this author in PubMed Google Scholar
Panke Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Shuo Huang
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

S.H., P.P.F, Z.Y.C., and S.Y.Z. conceived the project. Z.Y.C., and S.Y.Z. prepared the MspA nanopores. Z.Y.C., Y.Q.X. and W.D.J. performed the measurements. Y.Q.W., P.P.F. and Z.Y.C. designed the machine-learning and DBSCAN algorithms. P.K.Z. set up the instruments. S.H. P.P.F, and Z.Y.C. wrote the paper. S.H. supervised the project.

Corresponding author

Correspondence to Shuo Huang.

Ethics declarations

Competing interests

S.H., S.Y.Z. and Y.Q.W. have filed patents describing the preparation of heterogeneous MspA and its applications thereof. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of additional supplementary files

Supplementary Movie 1

Supplementary Movie 2

Supplementary Movie 3

Reporting Summary

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Fan, P., Cao, Z., Zhang, S. et al. Nanopore analysis of cis-diols in fruits. Nat Commun 15, 1969 (2024). https://doi.org/10.1038/s41467-024-46303-x

Download citation

Received: 09 January 2024
Accepted: 13 February 2024
Published: 05 March 2024
DOI: https://doi.org/10.1038/s41467-024-46303-x

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.