Integrated Size-Selective Cell Purification and Electroporation for Genetic Manipulation of Primary Cells

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents an integrated vortex-assisted microfluidic platform that combines size-selective cell trapping with electroporation and demonstrates improved processing capacity through electrode array redesign. The topic is relevant to the Micromachines readership, and the integration of upstream cell enrichment with intracellular delivery is potentially impactful for primary cell engineering. The experimental effort is substantial and the results are promising.

However, several critical engineering validation and reproducibility issues remain insufficiently addressed. The current version reads primarily as a functional demonstration rather than a rigorously validated microdevice study. The following major comments should be addressed before the manuscript can be considered for publication.

1. The redesigned electrode routing and scaling strategy is well presented and clearly improves electrical efficiency. However, it is not fully clear whether the local electric field distribution inside each chamber remains unchanged after scaling. Could the authors clarify how field uniformity across chambers was verified? To further strengthen the engineering validation, the authors may consider providing full-field simulations (e.g., COMSOL maps and field variation statistics) for the new device architecture.
2. The increase in processing flow rate is convincingly demonstrated. Still, I wonder whether the improved throughput also translates into a higher number of usable (viable and transfected) cells per experiment. The manuscript would benefit from reporting a functional output metric (e.g., viable transfected cells per run or effective processing yield) to better support the claim of practical processing capacity.
3. The optimization of buffer composition for primary cells is interesting and appears important for achieving successful delivery. Could the authors elaborate on why the Opti-MEM + DMSO condition improves transfection compared with DPBS-based buffers? Including a brief mechanistic analysis (such as conductivity, osmolarity, or membrane permeabilization considerations) would enhance reproducibility and scientific insight.
4. The observation of electrode degradation is informative and appreciated. It would be helpful to understand how this affects device usability during repeated operation. The authors may consider including repeated-use experiments and performance stability measurements to clarify device lifetime and practical applicability.
5. The comparison with chemical transfection provides useful biological context. I am curious how the proposed platform compares with other microfluidic electroporation approaches reported in the literature. A benchmark comparison table summarizing throughput, efficiency, and viability against representative microfluidic systems would help readers better appreciate the advancement of the device.

Author Response

Responses to Reviewer 1 comments

Comments 1: The redesigned electrode routing and scaling strategy is well presented and clearly improves electrical efficiency. However, it is not fully clear whether the local electric field distribution inside each chamber remains unchanged after scaling. Could the authors clarify how field uniformity across chambers was verified? To further strengthen the engineering validation, the authors may consider providing full-field simulations (e.g., COMSOL maps and field variation statistics) for the new device architecture.

Response 1: We thank the reviewer for this important engineering question. In the present design, the intra-chamber IDE geometry (strip width, gap spacing, and finger number) is identical to that of the previously validated 4×10 prototype (Ref. 28). In that work, COMSOL simulations were used to evaluate the local electric field magnitude at representative cell-orbit locations within the trapping chamber. Because the local electrode geometry remains unchanged in the scaled 12×12 architecture, the intra-chamber electric field distribution is governed by the same geometric parameters.

The primary modification in the present study concerns the routing architecture and chamber count. To verify preservation of field uniformity across the expanded array, we performed SPICE-based equivalent circuit modeling to quantify chamber-to-chamber voltage variability (Figure 2f). Since electric field magnitude within each chamber scales with the local voltage drop across the preserved IDE gap, maintaining voltage uniformity ensures preservation of per-chamber field conditions.

We have clarified this explanation in Section 3.1 (page 8, lines 328–353) of the revised manuscript. The added text reads as follows:

“Compared with planar parallel-plate configurations, the small inter-electrode gap in IDE geometries produces high local electric field strength at relatively low applied voltages. In the present device, vortex-trapped cells recirculate away from the electrode surface, occupying approximately the mid-to-upper portion of the microfluidic channel (~25 to 75% of the total channel height³¹), thereby remaining spatially separated from the electrode interface. Although electric field magnitude decreases with vertical distance from the electrode plane, the confined gap geometry enables sufficient field strength within this cell-trapping region to induce membrane polarization without requiring high global voltages across the full channel height. Because effective electroporation can be achieved at reduced applied voltage, overall current density, Joule heating, and electrochemical reactions at the electrode interface are minimized.

Additionally, the selected 20 µm electrode width balances field localization with fabrication and electrical stability. Maintaining sufficient exposed glass area between electrode features ensures robust PDMS-glass bonding during device assembly and prevents the common bus line from contacting the electrolyte, thereby avoiding unintended short-circuit pathways. This geometric balance supports both reliable electroporation performance and structural integrity during parallelized operation.

The intra-chamber IDE geometry (strip width, gap spacing, and finger number) was intentionally preserved from our previously validated 4×10 prototype²⁷. The number and lateral distribution of interdigitated electrode fingers within each chamber were designed to span the footprint of the vortex trapping region. By distributing multiple electrode pairs across the chamber area rather than concentrating them solely at the center, electric field exposure is maintained across both central and peripheral recirculation zones. This configuration helps ensure that cells orbiting along outer vortex trajectories experience sufficient field strength for membrane permeabilization.”

Comments 2: The increase in processing flow rate is convincingly demonstrated. Still, I wonder whether the improved throughput also translates into a higher number of usable (viable and transfected) cells per experiment. The manuscript would benefit from reporting a functional output metric (e.g., viable transfected cells per run or effective processing yield) to better support the claim of practical processing capacity.

Response 2: We thank the reviewer for this valuable suggestion. We agree that reporting a functional output metric strengthens the practical interpretation of processing capacity.

In response, we have added a quantitative estimate of effective processing yield in Section 3.4 (page 16, lines 649-652) of the revised manuscript. Specifically, we now report the number of viable, transfected primary fibroblasts obtained per experimental run by combining the measured trapping efficiency, recovery fraction, viability, and mRNA transfection efficiency. The added text reads as follows:

“When combined with measured trapping, recovery, and viability fractions, processing of a standard 4 mL sample (1000 cells/mL input concentration) yields an estimated ~170 viable mRNA-transfected primary fibroblasts per experimental run under optimized conditions.”

Comments 3: The optimization of buffer composition for primary cells is interesting and appears important for achieving successful delivery. Could the authors elaborate on why the Opti-MEM + DMSO condition improves transfection compared with DPBS-based buffers? Including a brief mechanistic analysis (such as conductivity, osmolarity, or membrane permeabilization considerations) would enhance reproducibility and scientific insight.

Response 3: We appreciate the reviewer’s request for additional mechanistic clarification regarding the improved performance of the Opti-MEM + DMSO formulation.

 

In the revised manuscript (Section 3.3, page 14, lines 579-585), we have expanded the discussion to provide a brief mechanistic interpretation of the observed buffer-dependent differences in transfection efficiency. Specifically, we now address the potential roles of buffer conductivity, osmolarity, and membrane-modifying additives in influencing electroporation dynamics. We emphasize that systematic physicochemical characterization (e.g., direct conductivity or osmolarity measurements under pulsing conditions) was beyond the scope of this study; however, the observed enhancement with Opti-MEM + DMSO is consistent with established electroporation principles involving conductivity-mediated field distribution and membrane stabilization effects. The added text reads as follows:

“The improved performance observed with the Opti-MEM + DMSO formulation likely reflects combined physicochemical effects. Buffer conductivity influences the effective electric field distribution across the inter-electrode gap and modulates induced transmembrane potential¹⁵. Differences in osmolarity may alter membrane tension and transient pore stability⁴³, thereby affecting cargo entry and post-pulse resealing dynamics. The reduced-serum composition of Opti-MEM may further mitigate post-electroporation stress and enhance recovery of primary cells.”

Comments 4: The observation of electrode degradation is informative and appreciated. It would be helpful to understand how this affects device usability during repeated operation. The authors may consider including repeated-use experiments and performance stability measurements to clarify device lifetime and practical applicability.

Response 4: We thank the reviewer for highlighting the importance of device durability and operational stability. The platform is intended for single-use applications in primary or patient-derived cell processing to minimize cross-contamination and ensure reproducibility. However, during immortalized cell testing, repeated electroporation experiments were conducted to assess operational stability. Specifically, the optimized 12×12 device was operated for up to 10 consecutive runs under standard voltage conditions without measurable changes in trapping efficiency, transfection performance, or viability. We have added clarification in the revised manuscript (Section 3.2, page 12, lines 505-510) describing this repeated-use evaluation. The added text reads:

“Although the device is intended for single-use operation in primary cell applications to minimize cross-contamination and ensure reproducibility, repeated electroporation experiments were conducted during immortalized cell testing. Stable trapping efficiency, voltage stability, and electroporation performance were maintained over up to 10 consecutive runs under standard operating conditions, indicating practical operational robustness in research use.”

Importantly, the reported electrode degradation (Figure S1) did not arise from cumulative usage but from the elevated voltage requirement of the intermediate 16×9 configuration. That design required input voltages exceeding 40 V to achieve the target electric field, resulting in high current density and electrochemical erosion of the microelectrode traces. In contrast, the optimized 12×12 configuration achieves the same field strength at substantially lower input voltage (~24 V), reducing current load and preserving electrode integrity under standard operating conditions. We have added clarification in Section 3.1 (page 10, lines 414-416) as follows:

“The degradation observed in the intermediate 16×9 configuration was therefore attributed to elevated voltage-induced current density rather than cumulative device usage.”

These clarifications distinguish voltage-induced material limitations from repeated-use effects and better define the practical operational limits of the device.

Comments 5: The comparison with chemical transfection provides useful biological context. I am curious how the proposed platform compares with other microfluidic electroporation approaches reported in the literature. A benchmark comparison table summarizing throughput, efficiency, and viability against representative microfluidic systems would help readers better appreciate the advancement of the device.

Response 5: We thank the reviewer for this constructive suggestion. In response, we have added a benchmark comparison table (Table 1) in the Discussion section (page 17, lines 716-719) summarizing representative microfluidic electroporation platforms across processing mode/reported capacity, sorting mechanism, integration level, applicable cell types, transfection efficiency, and viability.

Table 1 contextualizes the present platform relative to existing microfluidic electroporation systems and distinguishes delivery-only devices from integrated enrichment–electroporation architectures. This comparison clarifies the architectural advancement of the proposed system and better highlights its integration of size-selective trapping, localized IDE-based electroporation, and on-chip buffer exchange within a unified microfluidic workflow.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this article, the authors achieved a 5-fold increase in device processing throughput via the redesign of the electrode array architecture and electrical optimization, while preserving the size-selective cell trapping capability and stable electroporation performance of the platform. The research content of this manuscript is highly aligned with the scope and thematic focus of the Micromachines. This manuscript is acceptable for publication in this journal pending revisions, and a minor revision is hereby recommended. The authors are required to address the following specific comments point-by-point, and the manuscript will proceed to the re-review process upon completion and satisfactory revision.

1. It is recommended that the authors place each figure at the position where it is first mentioned in the main text.
2. The authors are recommended to supplement integrated sorting-electroporation experiments using simulated heterogeneous samples, to quantitatively characterize the sorting efficiency and enrichment purity of cells with different sizes, as well as the electroporation transfection efficiency and cell viability of the target cell population.
3. The authors are recommended to experimentally characterize the cell trapping efficiency of chambers at different positions within the array, quantify the performance variability across individual chambers, and provide data support for the stability of the device performance.
4. The authors are recommended to supplement a relevant performance comparison table in the Discussion section, to benchmark this work against previously published peer microfluidic studies across key dimensions including processing throughput, sorting mechanism, sorting resolution, applicable cell types, transfection efficiency, cell viability, and level of integration.

 

Author Response

Responses to Reviewer 2 comments

Comments 1: It is recommended that the authors place each figure at the position where it is first mentioned in the main text.

Response 1: Thank you for pointing this out, and we apologize for this oversight in the previous submission. We agree with this comment. Therefore, we have repositioned all figures in the revised manuscript so that each appears immediately after its first citation in the main text.

Comments 2: The authors are recommended to supplement integrated sorting-electroporation experiments using simulated heterogeneous samples, to quantitatively characterize the sorting efficiency and enrichment purity of cells with different sizes, as well as the electroporation transfection efficiency and cell viability of the target cell population.

Response 2: We thank the reviewer for this thoughtful suggestion. The vortex-based size-selective trapping module employed in this study has been previously validated using simulated heterogeneous spike-in samples, including enrichment of target cancer cells from mixed populations and whole blood matrices (Ref. 28). Importantly, the trapping chamber geometry and channel height remain unchanged in the present work; only the electrode routing architecture was redesigned to enable higher-volume operation.

To confirm preservation of size-selective trapping behavior in the scaled 12×12 array, we benchmarked trapping efficiency against the commercially validated VTX-1 configuration and quantified normalized per-chamber capture rates (Figure 2g, Figure S2). When normalized by chamber number, trapping performance remained comparable to the prior architecture, supporting retention of enrichment functionality under scaled operation.

In response to the reviewer’s comment, we have further clarified this point in the revised manuscript (Section 3.1, page 10, lines 432-435) by explicitly stating that the size-selective trapping geometry is unchanged from previously validated heterogeneous spike-in studies and that array-level benchmarking confirms preservation of enrichment performance. The added text reads as follows:

“Because the vortex trapping geometry and channel height are unchanged from previously validated heterogeneous spike-in enrichment studies^28,29, and per-chamber trapping efficiency remains comparable following array scaling, these results support preservation of enrichment functionality under integrated electroporation conditions.”

(Discussion, page 17, lines 712-715)

“The preservation of size-selective trapping behavior following electrode array scaling supports compatibility of the integrated architecture with previously demonstrated heterogeneous enrichment workflows.”

Comments 3: The authors are recommended to experimentally characterize the cell trapping efficiency of chambers at different positions within the array, quantify the performance variability across individual chambers, and provide data support for the stability of the device performance.

Response 3: Thank you for this thoughtful suggestion. We agree that chamber-to-chamber uniformity is important for scaled array operation.

In the current device architecture, all chambers operate in parallel and share a common outlet; cells are collected collectively following the trapping and electroporation steps. Because the system is designed for integrated continuous-flow operation rather than independent chamber addressing, experimental isolation of individual chamber outputs is not feasible without fundamentally altering the device configuration. As such, experimental measurements reflect aggregate array performance.

To evaluate chamber-to-chamber uniformity, we relied on electrical modeling and voltage distribution analysis using SPICE simulations (Figure 2f), which quantify inter-chamber voltage variability under scaled operation. Additionally, trapping efficiency normalized by chamber number was compared across configurations (Figure 2g and Figure S2), demonstrating preservation of size-selective trapping behavior at the array level.

We have clarified this in the revised manuscript (Section 3.1, page 9, lines 383-387):

“Because all chambers operate in parallel and share a common outlet within the integrated microfluidic architecture, experimental measurements reflect aggregate array behavior; chamber-to-chamber uniformity was therefore evaluated through electrical modeling and voltage variability analysis rather than independent chamber isolation.”

Comments 4: The authors are recommended to supplement a relevant performance comparison table in the Discussion section, to benchmark this work against previously published peer microfluidic studies across key dimensions including processing throughput, sorting mechanism, sorting resolution, applicable cell types, transfection efficiency, cell viability, and level of integration.

Response 4: Thank you for this valuable suggestion. We agree that comparative benchmarking enhances clarity and contextualization. In response, we have added a comparison table in the Discussion section (Table 1, page 17) summarizing representative microfluidic electroporation platforms across the reviewer-requested dimensions, including processing mode/reported capacity, sorting mechanism, level of integration, on-chip buffer exchange capability, electroporation configuration, applicable cell types, transfection efficiency, and viability. The table is structured to distinguish between delivery-only microfluidic electroporation systems and integrated cell-processing platforms. As shown in Table X, while several microfluidic systems demonstrate efficient intracellular delivery—particularly in homogeneous or pre-enriched suspensions—most do not incorporate upstream size-selective enrichment within the same microfluidic architecture. The present work uniquely integrates vortex-based size selection, parallelized localized electroporation, and on-chip buffer exchange within a unified trapping–electroporation workflow. This addition clarifies the architectural positioning and integration-level contribution of the current study relative to existing approaches.

The added text and table read as follows (Section 4, page 17, lines 716-728):

“To contextualize the present platform within the broader landscape of microfluidic electroporation technologies demonstrated in primary cells, representative systems are summarized in Table 1.

Table 1. Architectural and Functional Comparison of Representative Microfluidic Electroporation Platforms

Study	Sorting	Integration	Buffer Exchange	EP Format	Cell Type	Efficiency*	Viability*	Processing Mode
Huang 201721	None	Delivery	No	Planar (curved)	HUVEC	60–80% (DNA)	>80%	Flow (~1–2 mL/min)
Chang 202017	None	Delivery	No	Hybrid pulse (flow)	Primary T cells	Moderate–High	70–90%	10⁶–10⁷ cells/min
Lissandrello 202018	None	Delivery	No	Parallel flow	Primary T cells	~95% (mRNA)	>80%	~20M cells/min
Welch 202322	None	Delivery	No	RNP flow	Primary T cells	High (RNP)	70–78%	Clinical-scale flow
Duckert 202220	None	Delivery	No	Microelectrode array	Primary fibroblasts	60–80%	70–90%	Batch
Little 202323	None	Delivery	No	Droplet EP	Primary T cells	High	High	Droplet batch
This Work	Vortex (size-based)	Enrichment + EP	Yes	IDE array	Primary fibroblasts	76% (mRNA); 4–8% (DNA)	65–80%	Flow (5.2 mL/min)

*Reported as defined in the respective publications. Values for this work represent absolute re-porter-positive fractions and post-electroporation viability in primary human fibroblasts under optimized conditions.

As shown in Table 1, while several microfluidic electroporation platforms have demonstrated efficient intracellular delivery, including in certain primary cell types, most operate on pre-purified, homogeneous suspensions and focus exclusively on delivery. In contrast, the present work uniquely integrates size-selective enrichment, parallelized localized electroporation, and on-chip buffer exchange within a unified microfluidic architecture, enabling selective genetic manipulation of heterogeneous primary cell populations.”

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript shows a well-designed and carefully validated microfluidic platform that integrates size-selective cell trapping with scalable electroporation for primary cell transfection. The electrode array redesign, combined with systematic electrical and buffer optimization, enables increased throughput while preserving delivery performance. The extension from immortalized cell lines to primary human cells, together with comparative evaluation of plasmid size, mRNA delivery, and passage-dependent effects, strengthens the technical rigor and translational relevance of the study. Overall, the work is interesting and clearly written. However, several important issues related to mechanistic validation and functional assessment should be addressed to fully support the claims and enhance the impact of the manuscript. I therefore recommend major revision before the work can be considered for publication.

(1) The authors mentioned that " Chemical and viral gene delivery approaches introduce additional limitations, including reliance on carrier materials, cytotoxicity, immunogenicity, potential genome integration, and complex pre-selection workflows. Electroporation provides a non-viral and carrier-free alternative capable of delivering diverse cargoes, such as plasmids, mRNA, and proteins with performance tunable through electric field strength, pulse duration, and buffer composition." However, they have overlooked emerging nanostructure-induced intracellular delivery methods. For example, as reported in https://doi.org/10.1021/acsnano.2c07852, nanoneedle-based intracellular delivery demonstrates excellent delivery efficiency and cell viability, although it may be limited by fabrication complexity and challenges in system integration. The authors should cite some literature and discuss it in the Introduction section.

(2) Figure 1 appears to be missing. The authors seem to have overlooked it. Please check the manuscript carefully.

(3) Why are IDEs employed to generate localized electric fields? What is the advantage of IDEs compared with planar electrodes or others? The authors need to provide more underlying explanations and mechanisms.

(4)  How do the size and number of electrode strips in a single IDE affect electroporation? This also deserves experimental investigation and analysis in this work. When the width of the electrode strip is close to or even smaller than that of the cells, such as 10 μm or 5 μm, can it improve cell viability during electroporation?

(5) This work demonstrated effective intracellular delivery of plasmid DNA and mRNA. Cas9 protein&sgRNA (RNP) can function directly without the need for gene expression, emerging as a popular strategy. For example, as reported in https://doi.org/10.1021/acsnano.2c07852, nanoneedle-based intracellular delivery methods can deliver Cas9 protein&sgRNA (RNP) directly. It is recommended that experimental demonstrations of protein delivery using this electroporation method be provided. At the very least, some discussion or a future outlook related to protein delivery should be included.

Author Response

Responses to Reviewer 3 comments

Comments 1: The authors mentioned that " Chemical and viral gene delivery approaches introduce additional limitations, including reliance on carrier materials, cytotoxicity, immunogenicity, potential genome integration, and complex pre-selection workflows. Electroporation provides a non-viral and carrier-free alternative capable of delivering diverse cargoes, such as plasmids, mRNA, and proteins with performance tunable through electric field strength, pulse duration, and buffer composition." However, they have overlooked emerging nanostructure-induced intracellular delivery methods. For example, as reported in https://doi.org/10.1021/acsnano.2c07852, nanoneedle-based intracellular delivery demonstrates excellent delivery efficiency and cell viability, although it may be limited by fabrication complexity and challenges in system integration. The authors should cite some literature and discuss it in the Introduction section.

Response 1: Thank you for pointing this out. We agree with this comment and appreciate the suggestion to broaden the contextual discussion of emerging intracellular delivery technologies. Therefore, we have revised the Introduction (Section 1, page 1, lines 36-60) to incorporate discussion of nanostructure-mediated and mechanically assisted intracellular delivery approaches alongside electroporation. Specifically, we have reorganized the paragraph structure to first introduce physical intracellular delivery strategies as a broader category, followed by mechanical approaches and then electroporation, before narrowing to microscale electroporation platforms. In this revision, we have added citations to nanoneedle-based platforms (ACS Nano 2022), cell-squeezing approaches (Sharei et al., PNAS 2013), and recent high-throughput 2D mechanoporation systems (Adv. Healthcare Mater., 2025). The modified introduction section reads as follow:

 “As alternatives, physical intracellular delivery strategies have been developed to facilitate membrane permeabilization without exogenous carriers.

Mechanically assisted intracellular delivery approaches, including nanoneedle-based platforms⁸, cell-squeezing approaches⁹, and recent high-throughput 2D mechanoporation systems¹⁰, induce membrane disruption through localized deformation. These systems have demonstrated efficient intracellular delivery with preserved cell viability. However, challenges related to device fabrication complexity, substrate dependence, and integration within heterogeneous sample-processing workflows remain active areas of investigation.

Electroporation represents another widely adopted physical delivery strategy, enabling transient membrane permeabilization through externally applied electric fields. Performance can be tuned through electric field strength, pulse duration, and buffer composition^4,5,11–14. Nevertheless, conventional bulk electroporation systems often require high voltages and may generate spatially non-uniform electric fields, resulting in variable membrane permeabilization and reduced reproducibility, particularly when applied to heterogeneous primary cell populations^14,15.

Microscale electroporation platforms implemented within microfluidic architectures have addressed several of these limitations by enabling localized field control, precise fluid handling, and parallelized operation^16–19. Designs including microelectrode arrays²⁰, continuous flow-through electroporation systems^18,19,21,22, and droplet-based formats^23,24, have demonstrated efficient gene delivery to primary cells. However, most focus exclusively on intracellular delivery and assume pre-purified, tightly size-distributed input populations, leaving upstream cell isolation and downstream electroporation as separate processes.”

Comments 2: Figure 1 appears to be missing. The authors seem to have overlooked it. Please check the manuscript carefully.

Response 2: We sincerely apologize for this oversight in the previous submission. Thank you for bringing this to our attention. Figure 1 was inadvertently omitted during manuscript formatting. We have now restored Figure 1 and ensured that it appears at its first citation in the main text (page 2, lines 86-99) in the revised manuscript. In addition, we have carefully reviewed the entire manuscript to confirm that all figures are correctly included, labeled, and referenced.

Comments 3: Why are IDEs employed to generate localized electric fields? What is the advantage of IDEs compared with planar electrodes or others? The authors need to provide more underlying explanations and mechanisms.

Response 3: Thank you for this important and insightful comment. We agree that the rationale for selecting interdigitated electrodes (IDEs) and the associated physical mechanisms should be clarified.

In response, we have added a detailed mechanistic explanation in Section 3.1 (page 8, lines 328-345) describing the advantages of IDE configurations over planar parallel-plate electrodes. Specifically, we now explain that the small inter-electrode gap in IDE geometries produces high local electric field strength at relatively low applied voltages. Because induced transmembrane potential scales with local electric field magnitude, this confined-gap configuration enables effective membrane polarization within the vortex-trapping region without requiring high global voltages across the full channel height.

 

We further clarify that in the present device, vortex-trapped cells recirculate within the mid-to-upper portion of the channel height (~25–75% of total channel height), remaining spatially separated from the electrode interface. Although electric field magnitude decreases with vertical distance from the electrode plane, the localized IDE geometry maintains sufficient field strength at the cell-orbit region while minimizing overall current density, Joule heating, and electrochemical reactions.

In addition, we discuss fabrication and electrical stability considerations: the selected 20 µm electrode width preserves exposed glass area to ensure robust PDMS–glass bonding and prevents the common bus line from contacting the electrolyte, thereby avoiding unintended short-circuit pathways. These combined factors motivated the selection of IDEs for scalable, parallelized operation within the integrated trapping–electroporation architecture. The added text in the manuscript reads as follows:

“Compared with planar parallel-plate configurations, the small inter-electrode gap in IDE geometries produces high local electric field strength at relatively low applied voltages. In the present device, vortex-trapped cells recirculate away from the electrode surface, occupying approximately the mid-to-upper portion of the microfluidic channel (~25 to 75% of the total channel height³¹), thereby remaining spatially separated from the electrode interface. Although electric field magnitude decreases with vertical distance from the electrode plane, the confined gap geometry enables sufficient field strength within this cell-trapping region to induce membrane polarization without requiring high global voltages across the full channel height. Because effective electroporation can be achieved at reduced applied voltage, overall current density, Joule heating, and electrochemical reactions at the electrode interface are minimized.

Additionally, the selected 20 µm electrode width balances field localization with fabrication and electrical stability. Maintaining sufficient exposed glass area between electrode features ensures robust PDMS-glass bonding during device assembly and prevents the common bus line from contacting the electrolyte, thereby avoiding unintended short-circuit pathways. This geometric balance supports both reliable electroporation performance and structural integrity during parallelized operation.”

Comments 4: How do the size and number of electrode strips in a single IDE affect electroporation? This also deserves experimental investigation and analysis in this work. When the width of the electrode strip is close to or even smaller than that of the cells, such as 10 μm or 5 μm, can it improve cell viability during electroporation?

Response 4: We thank the reviewer for this insightful question regarding electrode geometry scaling. The influence of interdigitated electrode (IDE) strip dimensions and finger number on electric field distribution was systematically investigated in our previously reported 4×10 vortex-assisted electroporation prototype (Ref 28), where COMSOL simulations were used to evaluate achievable local electric field magnitude at representative cell-orbit locations within the trapping chamber. Based on those studies, a 20 μm electrode width and corresponding gap spacing were selected as a balance between effective field localization, spatial uniformity, and fabrication robustness.

In the present work, the intra-chamber IDE geometry (strip width, gap spacing, and number of fingers) was intentionally preserved to maintain the validated per-chamber electroporation environment. The focus of this study was to redesign the electrode routing architecture to increase the number of chambers and improve voltage efficiency, thereby enhancing array-level processing capacity while maintaining established electroporation performance.

While reducing strip width may increase local electric field gradients near electrode edges, excessively narrow features amplify edge-concentrated fields and electrochemical activity, which may increase membrane stress rather than improve viability. In addition, features below ~10 μm approach practical resolution limits of the photolithography and lift-off process used for 300 nm Au electrodes and may compromise bonding integrity and electrical stability.

We have clarified this scope and rationale in Section 3.1 (page 8, lines 346-353) of the revised manuscript. The added text reads:

“The intra-chamber IDE geometry (strip width, gap spacing, and finger number) was intentionally preserved from our previously validated 4×10 prototype²⁷. The number and lateral distribution of interdigitated electrode fingers within each chamber were designed to span the footprint of the vortex trapping region. By distributing multiple electrode pairs across the chamber area rather than concentrating them solely at the center, electric field exposure is maintained across both central and peripheral recirculation zones. This configuration helps ensure that cells orbiting along outer vortex trajectories experience sufficient field strength for membrane permeabilization.”

Comments 5: This work demonstrated effective intracellular delivery of plasmid DNA and mRNA. Cas9 protein&sgRNA (RNP) can function directly without the need for gene expression, emerging as a popular strategy. For example, as reported in https://doi.org/10.1021/acsnano.2c07852, nanoneedle-based intracellular delivery methods can deliver Cas9 protein&sgRNA (RNP) directly. It is recommended that experimental demonstrations of protein delivery using this electroporation method be provided. At the very least, some discussion or a future outlook related to protein delivery should be included.

Response 5: We thank the reviewer for this valuable suggestion. Direct protein delivery, including Cas9 ribonucleoprotein (RNP) complexes, is indeed an important and emerging strategy for gene editing applications. In our previously reported low-throughput vortex-assisted electroporation platform (Refs. 22–24), we demonstrated sequential multi-molecule delivery, including protein cargos, using the same interdigitated electrode chamber geometry.

In the present study, the intra-chamber electroporation configuration (strip width, gap spacing, and electrode arrangement) remains unchanged, while architectural scaling was achieved through routing and array optimization. Therefore, the current platform design is consistent with previously demonstrated protein delivery capability, and extension to protein and RNP delivery represents a logical future direction.

Although the primary focus of this manuscript is architectural scaling, electrical optimization, and validation in primary human cells, we have added discussion (page 17, lines 696-704) in the revised manuscript outlining the applicability of the platform to protein and RNP delivery and positioning this as an important future direction for translational genome engineering applications. The added text reads:

“In addition to plasmid DNA and mRNA delivery, direct protein and ribonucleoprotein (RNP) delivery represent important strategies for genome editing applications. In our previously reported vortex-assisted electroporation platforms, sequential multi-molecule delivery including protein cargos was demonstrated using the same interdigitated electrode chamber geometry²⁷. Because the present work preserves the local electroporation configuration, the platform design remains consistent with previously demonstrated protein delivery capabilities and may be extended to RNP delivery. Future studies will evaluate CRISPR-Cas9 RNP delivery within the integrated trapping-electroporation workflow to further expand the platform’s applicability in primary cell engineering.”

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for your careful revisions and for the detailed responses to the reviewer comments. The explanations provided have clarified the main concerns raised during the previous review. In particular, the clarification of the electrode routing strategy and the explanation regarding preservation of the intra-chamber IDE geometry help to better justify the scalability of the device architecture. The addition of a quantitative estimate of the effective processing yield also strengthens the practical interpretation of the system’s performance. I also appreciate the expanded discussion on the Opti-MEM + DMSO buffer formulation and the clarification regarding device durability and voltage-induced degradation. These additions improve the overall clarity and completeness of the manuscript.

Overall, the revisions have improved the manuscript and adequately addressed the reviewer’s comments. I have no further major concerns and believe the manuscript is suitable for publication.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have clearly addressed the reviewers' concerns. Acceptance is now recommended.