Scalability of U-Shape Magnetic Nanoparticles-Based Microreactor–Lipase-Catalyzed Preparative Scale Kinetic Resolutions of Drug-like Fragments

Round 1

Reviewer 1 Report

This work investigated the scalability of a tunable U-shape magnetic nanoparticles (MNPs)-based microreactor for the catalytic application of Lipase-catalyzed kinetic resolutions. The microreactors with various tube sizes and MNP capacities were tested and the kinetic resolution of the drug-like alcohols 4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-ol (±) and 4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-ol (±)-1b utilizing Lipase B from Candida antarctica immobilized covalently onto MNPs leading to highly enantioenriched products. This work is an extension from a previous study, Catalysts 2022, 12, 1065, a systematic study, however, lack of novelty since the reaction, catalyst, substrate, reactor are the same. I suggest rejecting this paper and the authors should submit to a lower impact journal. 

 

1. The significance of this work is scalability. The authors claimed a preparative scale catalysis. However, based on the SI, the scale of the reaction is ~400 mg, which is far from preparative scale in pham synthesis. 

 

2. Also, for a synthesis in a preparative scale, I suggest reporting isolated yield. 

 

3. I am very curious if there is a particular reason why the author designed the reactors as U-shape? Why not, such as linear, or wave shape (more U)?

 

4. Please add error bar to Figure 1 and Figure 3.

 

5. I suggest the author perform simulation based on Figure 4 for a better explanation.

 

6. I suggest the author further investigating the leaching of Fe NP, such as ICP-MS for the reaction solution. 

 

7. I also suggest the authors test the ability to recycle and reuse the catalyst.

Author Response

Comments and Suggestions for Authors

This work investigated the scalability of a tunable U-shape magnetic nanoparticles (MNPs)-based microreactor for the catalytic application of Lipase-catalyzed kinetic resolutions. The microreactors with various tube sizes and MNP capacities were tested and the kinetic resolution of the drug-like alcohols 4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-ol (±) and 4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-ol (±)-1b utilizing Lipase B from Candida antarctica immobilized covalently onto MNPs leading to highly enantioenriched products. This work is an extension from a previous study, Catalysts 2022, 12, 1065, a systematic study, however, lack of novelty since the reaction, catalyst, substrate, reactor are the same. I suggest rejecting this paper and the authors should submit to a lower impact journal. 

 

Response: Our aim with this study was to investigate the effect of tube diameter on the behavior of the tubular MNP reactor, therefore the reactor is not essentially the same. Although the catalyst is really the same as applied in the work published Catalysts 2022, 12, 1065—for the sake of comparability—the substrates are new. Although the racemic forms of the drug-like alcohols 4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-ol (±)-1a and 4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-ol (±)-1b are already described in the literature, their pure enantiomers were never prepared before. The same is true for the produced (R)-4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-yl acetate (R)-2a and (R)-4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-yl acetate (R)-2b.

 

1. The significance of this work is scalability. The authors claimed a preparative scale catalysis. However, based on the SI, the scale of the reaction is ~400 mg, which is far from preparative scale in pham synthesis. 

 

Response: As picked up, a major topic of this work was the investigation of optimal tube diameter for such type of reactors. The first studies—using novel KRs of (±)-1a and (±)-1b as model reactions—were logically made on small scale (several 10-100 mg). Further scaling up is obviously possible by numbering up the cells (up to 100 without significant problem) and parallelization of the multicell devices. These are the planned topics of our forthcoming studies.

 

2. Also, for a synthesis in a preparative scale, I suggest reporting isolated yield. 

 

Response: In our revised manuscript, additional data on isolated yields are given (Table 2).

 

3. I am very curious if there is a particular reason why the author designed the reactors as U-shape? Why not, such as linear, or wave shape (more U)?

 

Response: As we mentioned in our response to point 1., numbering up the cells seems easily implementable. To make this pint clear we have added the following paragraph to the revised manuscript: “It is considered that after finding the optimal tube diameter, numbering up the cells seems easily implementable. A logical setup is to use the wave shape (more of Us) reactor setup. For example, 9 turns with 10 cells after each turn would represent a 100-cell reactor holding up to 3.6 g of MNP biocatalyst within a less than 2 m-long tube fitting onto a less than 20 cm x 20 cm plate. Such plate would represent a multigram-scale reactor; and such gram scale reactors are easily parallelizable with standard joining elements. The U-shape reactor in this study approaches the first part of this geometry.”

 

4. Please add error bar to Figure 1 and Figure 3.

 

Response: The batch mode experiments were preliminary studies for the flow reactor investigations. Therefore, the batch reactions were performed as single experiments. As we observed no outliers in the progress curves, and the trends corresponded to the expectations, we thought no reason to repeat those experiments. The error bars were added to Figure 3.

 

5. I suggest the author perform simulation based on Figure 4 for a better explanation.

 

Response: Another further aim is to model the situation which is shown schematically in Figure 4. This work, however, is not quite easy and requires modelling of the MNPs packing in the magnetic field considering the particle size. Modeling the microenvironment close to the MNPs surface with different viscosity as in the bulk fluid is also not obvious. This modeling—in co-operation with experts in that field—is also among our forthcoming projects. Studies, like the present one, can provide experimental data comprising frame and input data for modelling works.

 

6. I suggest the author further investigating the leaching of Fe NP, such as ICP-MS for the reaction solution. 

 

Response: Our setups contained a simple visual detection (an empty transparent chamber after the outlet of the six-chamber reactor with bright white background and another permanent magnet). In this way, accumulation of the leaching MNPs become visible when happened.

 

7. I also suggest the authors test the ability to recycle and reuse the catalyst.

 

Response: The revised manuscript contains significant extensions on recycling and reusing the CaLB-MNP biocatalyst, focusing on the continuous-flow setup.

Reviewer 2 Report

This manuscript reports the immobilization of lipase on magnetic nanoparticles and applying them in a U-shape microreactor for continuous flow kinetic resolutions. In general, most experiments were carefully performed, the results were clearly presented, and the material and methods were well documented. However, the authors just reported another very similar project (Ref 55): A Convenient U-Shape Microreactor for Continuous Flow Biocatalysis with Enzyme-Coated Magnetic Nanoparticles-Lipase-Catalyzed Enantiomer Selective Acylation of 4-(Morpholin-4-yl)butan-2-ol. Catalysts 2022, 12, 1065. This manuscript is very similar to that previous report: the same catalysts (lipase on MNPs), the same reaction set-up (U-shape microreactor), and very similar substrates (both are racemic N-substituted butan-2-ol). I did not see any significant advances compared to the previous report. The authors should provide more new experiments and new data to justify the publication of the current manuscript. Other comments: 

1. Introduction: too lengthy, I suggest the authors shorten the introduction and make the manuscript more concise. 
2. Results: The authors performed the kinetic resolutions in batch mode and continuous mode, however, there is not a good comparison of these two modes. What are the pros and cons of continuous mode? Please discuss these using the experimental data. 
3. Results: The authors claimed the “scalability” of the process, however, they only performed the preparative scale kinetic resolution in batch mode. To demonstrate the scalability of the U-shape microreactor, the preparative scale synthesis is necessary and better on a large scale (e.g., at least a gram scale). 
4. Typos: Line 146 “sp3” should be “sp2”. Line 289 “1a” should be “1b”. 

Overall, the manuscript is too similar to the previous publication (Catalysts 2022, 12, 1065), the authors should provide enough new data and new technical/scientific insights to justify the publication of the current report. 

Author Response

Comments and Suggestions for Authors

This manuscript reports the immobilization of lipase on magnetic nanoparticles and applying them in a U-shape microreactor for continuous flow kinetic resolutions. In general, most experiments were carefully performed, the results were clearly presented, and the material and methods were well documented. However, the authors just reported another very similar project (Ref 55): A Convenient U-Shape Microreactor for Continuous Flow Biocatalysis with Enzyme-Coated Magnetic Nanoparticles-Lipase-Catalyzed Enantiomer Selective Acylation of 4-(Morpholin-4-yl)butan-2-ol. Catalysts 2022, 12, 1065. This manuscript is very similar to that previous report: the same catalysts (lipase on MNPs), the same reaction set-up (U-shape microreactor), and very similar substrates (both are racemic N-substituted butan-2-ol). I did not see any significant advances compared to the previous report. The authors should provide more new experiments and new data to justify the publication of the current manuscript. Other comments:

 

Response: Our aim with this study was to investigate the effect of tube diameter on the behavior of the tubular MNP reactor, therefor the reactor is not essentially the same. Although the catalyst is really the same as applied in the work published Catalysts 2022, 12, 1065—for the sake of comparability—the substrates are new. Although the racemic forms of the drug-like alcohols 4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-ol (±)-1a and 4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-ol (±)-1b are already described in the literature, their pure enantiomers were never prepared before. The same is true for the produced (R)-4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-yl acetate (R)-2a and (R)-4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-yl acetate (R)-1b. We feel that our present communication provides real novelties as compared to the previous paper.

 

1. Introduction: too lengthy, I suggest the authors shorten the introduction and make the manuscript more concise. 

 

Response: The Introduction should provide background to several aspects of this work. One easy way would be to remove those parts which were introduced in our previous work appeared in Catalysts 2022, 12, 1065. However, in this case the paper would not be easily readable alone without having the previous article. Therefore, we think that the Introduction should contain background for using biocatalysis and especially CaLB, background to selection of the substrates, background on enzyme immobilization onto MNPs and the background of MNP-based bioreactors. Nevertheless, we shortened the Introduction by ~11% (to 2800 words from 3150 words).

 

2. Results: The authors performed the kinetic resolutions in batch mode and continuous mode, however, there is not a good comparison of these two modes. What are the pros and cons of continuous mode? Please discuss these using the experimental data.

 

Response: We have added space time yield data for the two modes as a comparison.  

 

3. Results: The authors claimed the “scalability” of the process, however, they only performed the preparative scale kinetic resolution in batch mode. To demonstrate the scalability of the U-shape microreactor, the preparative scale synthesis is necessary and better on a large scale (e.g., at least a gram scale).
   
   

Response: As we mentioned in our response to Referee 1 as well, numbering up the cells seems easily implementable. To make this pint clear we have added the following paragraph to the revised manuscript: “It is considered that after finding the optimal tube diameter, numbering up the cells seems easily implementable. A logical setup is to use the wave shape (more of Us) reactor setup. For example, 9 turns with 10 cells after each turn would represent a 100-cell reactor holding up to 3.6 g of MNP biocatalyst within a less than 2 m-long tube fitting onto a less than 20 cm x 20 cm plate. Such plate would represent a gram-scale reactor; and such gram scale reactors are easily parallelizable with standard joining elements. The U-shape reactor in this study approaches the first part of this geometry.”

 

4. Typos: Line 146 “sp3” should be “sp2”. Line 289 “1a” should be “1b”. 

 

Response: The typos have been corrected. Thank you for improving our manuscript by your comments.

 

Overall, the manuscript is too similar to the previous publication (Catalysts 2022, 12, 1065), the authors should provide enough new data and new technical/scientific insights to justify the publication of the current report.

 

Response: As pointed out in our response to your note 1, our aim with this study was to investigate the effect of tube diameter on the behavior of the tubular MNP reactor, therefore the reactor is developed further. Moreover, the enantiomeric forms of the drug-like alcohols 4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-ol 1a and 4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-ol 1b as well as their acetates (R)-2a and (R)- (R)-2b were never prepared before. In addition, the revised manuscript contains significant extensions on recycling and reusing the CaLB-MNP biocatalyst, focusing on the continuous-flow setup. Thus, we feel that our present communication provides real novelties as compared to the previous paper.

Reviewer 3 Report

The article evaluates the scalability of a novel U-shape flow reactor with magnetic nanoparticles coated with CaLB in the kinetic resolutions of 4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-ol and 4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-ol. This manuscript is well written, with a very detailed introduction. This work can be seen as a continuation of a previously work published also in Catalyst by the same group (10.3390/catal12091065).

 

I have some major and minor comments/questions:

Major:

1. The authors suggest that presented results are promising considering the scalability of the method. I agree that seeing a good performance with with bigger ID tubing (1.5mm) sounds promising but I would suggest to test long term operational stability of the reactor, similarly as performed in the previous Catalyst report (10.3390/catal12091065) Also, it would be more appealing for the manuscript to include a longer protocol to prepare products around g scale, and not just mg scale.

Minor:

1. Line 217. Why do the authors use preparative TLC? It sounds contradictory to me if a method to make KR in bigger scale is developed, to use such a purification method. Can be a different method used? Have the authors tried it?

2. Figure 2 and Line 272. The CSA of a 0.75 mm tubing is 0.44 mm².

3. Leaching is firstly mentioned in line 271 but not defined until line 315. I suggest to define this before in the text to help reading flow.

4. Figure 4: How this images were obtained? Are they some class of simulation or just a schematic view of what authors propose to explain the observations? This should be clarified. Also, what are those rectangles above (a), (b) and (c)? No explanation of those is written and those three look exactly the same.

5. What kind of pumps are used? What brand? What is the maximum pressure tolerated? This information is very important to add in order to guarantee reproducibility of the experiments.

6. Supporting information file: What is the source of 1,2,3,4-tetrahydroisoquinoline and 1,2,3,4-tetrahydroquinoline starting materials? This informations is missing in Materials section.

7. Supporting information file: Characterization of 1a: Figure S3 (¹³C-NMR 1a) shows one peak more than peaks report (Section 3.3, Page 3). The peak at 133.75 ppm should be added.

8. Supporting Information file: 1H-NMRs in Figures S2 and S6 show a ~5.25 ppm peak. What is this peak assigned to?

I consider this article is acceptable for publication after suggested revisions are met and that it has a good fit with the journal scope.

 

 

Author Response

Comments and Suggestions for Authors

The article evaluates the scalability of a novel U-shape flow reactor with magnetic nanoparticles coated with CaLB in the kinetic resolutions of 4-(3,4-dihydroisoquinolin-2(1H)-yl)butan-2-ol and 4-(3,4-dihydroquinolin-1(2H)-yl)butan-2-ol. This manuscript is well written, with a very detailed introduction. This work can be seen as a continuation of a previously work published also in Catalyst by the same group (10.3390/catal12091065).

 

Response: Thank you for the positive opinion.

 

I have some major and minor comments/questions:

 

Major:

1. The authors suggest that presented results are promising considering the scalability of the method. I agree that seeing a good performance with with bigger ID tubing (1.5 mm) sounds promising but I would suggest to test long term operational stability of the reactor, similarly as performed in the previous Catalyst report (10.3390/catal12091065) Also, it would be more appealing for the manuscript to include a longer protocol to prepare products around g scale, and not just mg scale.

 

Response: As detailed in our responses to the two other reviewers, we extended our manuscript on further considerations on upscaling. In addition, the revised manuscript contains significant extensions on recycling and reusing the CaLB-MNP biocatalyst, focusing on the continuous-flow setup.

 

Minor:

1. Line 217. Why do the authors use preparative TLC? It sounds contradictory to me if a method to make KR in bigger scale is developed, to use such a purification method. Can be a different method used? Have the authors tried it?

 

Response: The major topic of this work was the investigation of optimal tube diameter for such type of reactors. The first studies—using novel KRs of (±)-1a and (±)-1b as model reactions—were logically made on small scale (several 10-100 mg). The TLC separations selected for this scale.

Considerations for further possibilities of upscaling are discussed in a new section: “It is considered that after finding the optimal tube diameter, numbering up the cells seems easily implementable. A logical setup is to use the wave shape (more of Us) reactor setup. For example, 9 turns with 10 cells after each turn would represent a 100-cell reactor holding up to 3.6 g of MNP biocatalyst within a less than 2 m-long tube fitting onto a less than 20 cm x 20 cm plate. Such plate would represent a gram-scale reactor; and such gram scale reactors are easily parallelizable with standard joining elements. The U-shape reactor in this study approaches the first part of this geometry.”

 

2. Figure 2 and Line 272. The CSA of a 0.75 mm tubing is 0.44 mm².

 

Response: CSA of a 0.75 mm tube in Figure 2 and in the text is corrected to 0.44 mm².

 

3. Leaching is firstly mentioned in line 271 but not defined until line 315. I suggest to define this before in the text to help reading flow.

 

Response: “Leaching” has been defined at the first occasion.

 

4. Figure 4: How this images were obtained? Are they some class of simulation or just a schematic view of what authors propose to explain the observations? This should be clarified. Also, what are those rectangles above (a), (b) and (c)? No explanation of those is written and those three look exactly the same.

 

Response: We extended the legend for Figure 4: … Schematic behavior of a CaLB-MNP-containing reaction cell in a tubular MNP flow reactor with tubes of different inner diameter (the rectangles below the tubes of various diameters show the polarity and magnetic field withing the same permanent magnet).

 

5. What kind of pumps are used? What brand? What is the maximum pressure tolerated? This information is very important to add in order to guarantee reproducibility of the experiments.

 

Response: We have added a part on the pump: “A SpinSplit continuous flow syringe pump (SpinSplit Technical Research and Development LLC, Budapest, Hungary) equipped with two glass syringes of 0-5 mL volume was coupled to tubular reactor part of the reactor module.” During our work, no problems due to hydrostatic resistance have been observed.

 

6. Supporting information file: What is the source of 1,2,3,4-tetrahydroisoquinoline and 1,2,3,4-tetrahydroquinoline starting materials? This informations is missing in Materials section.

 

Response: The supplementary information file is completed with the required information: “… 1,2,3,4-tetrahydroquinoline, 1,2,3,4-tetrahydroisoquinoline, and vinyl acetate were purchased from Merck (Darmstadt, Germany) or Alfa Aesar Europe (Karlsuhe, Germany).”

 

7. Supporting information file: Characterization of 1a: Figure S3 (¹³C-NMR 1a) shows one peak more than peaks report (Section 3.3, Page 3). The peak at 133.75 ppm should be added.

 

Response: A peak at 133.8 has been added to the text. Thank you.

 

8. Supporting Information file: 1H-NMRs in Figures S2 and S6 show a ~5.25 ppm peak. What is this peak assigned to?

 

Response: The residual dichloromethane peaks are assigned in the SI.

 

I consider this article is acceptable for publication after suggested revisions are met and that it has a good fit with the journal scope.

 

Response: Thank you for your opinion. We feel that our present communication providing real novelties as compared to the previous paper will be eligible for publication in Catalysts.

Round 2

Reviewer 1 Report

I would like to thank the authors for making efforts to address my comments. Unfortunately, I was still not convinced that it was novel enough to be published again in Catalysts. I suggest the editors looking for a second reviewer or rejecting this paper.

Reviewer 2 Report

The authors had performed additional experiment and simulation, carefully revised the manuscript, and well addressed all my questions. The revised manuscript is much improved now and suitable for publication.