Controlling Assembly of Hybrid DNA Nanostructures into Higher‐Order Structures via Hydrophobicity

Abstract

Hydrophobic interactions are one of the fundamental driving forces of self‐assembly in living systems. It remains challenging to harness hydrophobicity to have a controllable and programmable assembly of DNA nanostructures. On the other hand, there is also a need to explore orthogonal hierarchical assembly strategies to be used as an additional toolset along with the traditional Watson–Crick base pairing to achieve complex superstructures. In this work, we rationally design and synthesize a series of low molecular weight hydrophobic molecules that are conjugated to single‐stranded DNA strands. By incorporating these modified DNA strands into the precisely defined locations of DNA tiles and origami nanostructures, we achieve controlled hierarchical assembly driven by hydrophobic interaction. We demonstrate a versatile hydrophobicity‐guided higher‐order assembly strategy by employing strategically engineered DNA nanostructures of increasing complexity, ranging from simple DNA tiles to complex origami structures, functionalized with these small hydrophobic molecules as programmable building blocks.

We harness hydrophobic interactions, fundamental in biology yet underutilized in DNA nanotechnology, to build a design framework for hierarchical DNA assemblies. By conjugating small hydrophobic molecules to DNA motifs, we controlled the formation of higher‐order structures, moving beyond traditional Watson–Crick pairing.

Introduction

Hydrophobic interactions play important roles in self‐assembly processes in biological systems, driving the organization and stability of macromolecules in aqueous environments. These interactions arise when nonpolar molecules or regions aggregate to minimize their contact with water, leading to a primarily entropy‐driven stabilization.^{[ 1} , ^{2 ]} In protein folding, hydrophobic amino acid residues cluster in the core, stabilizing the native structure and influencing function.^{[ 3} , ⁴ , ^{5 ]} Similarly, lipid bilayer formation in cell membranes relies on hydrophobic forces to assemble phospholipids, with nonpolar tails facing inward and hydrophilic heads outward, creating selective permeability.^{[ 6} , ^{7 ]} In molecular recognition and supramolecular assembly, hydrophobic regions promote specific binding, crucial for enzyme‐substrate interactions^{[ 8} , ^{9 ]} and signal transduction.^{[ 10} , ^{11 ]} Additionally, hydrophobic interactions contribute to the self‐assembly of nanoscale biological structures like micelles^{[ 12 ]} and viral capsids.^{[ 13 ]} Misfolding and aggregation may lead to diseases such as Alzheimer's^{[ 14 ]} and Parkinson's,^{[ 15 ]} showing the importance of maintaining biological order and function. Hence, it is important to study hydrophobic interactions as a way to achieve controllable and programmable self‐assembly.

The rapid growth of DNA nanotechnology over the past few decades has yielded well‐defined DNA nanoarchitectures with diverse shapes, sizes, and structural complexity, showing great potential in materials engineering and biomedical applications.^{[ 16 ]} The unique addressability of DNA nanostructures makes them ideal model systems for introducing functional entities at prescribed locations with sub‐nanometer precision. Examples of such modifications range from dyes,^{[ 17} , ^{18 ]} quantum dots,^{[ 19 ]} polymers,^{[ 20 ]} and peptides^{[ 21 ]} to enzymes.^{[ 22} , ^{23 ]} Previous studies have demonstrated that hydrophobically modified synthetic DNA oligos, incorporating moieties such as cholesterol, porphyrin, and azobenzene, can spontaneously self‐assemble into micelles, vesicles, and 2D lamellar sheets in aqueous medium due to their amphiphilic properties.^{[ 24} , ^{25 ]} With advancements in DNA nanotechnology, integrating hydrophobic modifications into highly programmable DNA architectures has been leveraged for enhanced gene and drug delivery,^{[ 26} , ^{27 ]} protein assembly,^{[ 28} , ^{29 ]} and artificial membrane channel formation.^{[ 30} , ^{31 ]} Notable examples of hydrophobicity‐driven DNA assemblies include the geometry‐dependent self‐assembly of DNA cages enabled by site‐specific positioning of dendritic alkyl chains,^{[ 32 ]} quantized aggregation of DNA nanocages via sequence‐defined hydrophobic polymers,^{[ 33 ]} exclusive tetramerization of DNA nanostructures through blunt‐end cholesterol modifications,^{[ 34 ]} and higher‐order assembly guided by perfluorinated tags incorporated into 2D origami edges.^{[ 35 ]} Responsive hybrid nanostructures have also been recently demonstrated in the form “giant surfactants” created by functionalizing DNA tetrahedron with temperature‐responsive polymers^{[ 36 ]} and the reversible reconfiguration of a rectangular DNA origami functionalized with elastin‐like polypeptides.^{[ 37 ]} However, there still exists only a limited number of such systems integrating hydrophobic molecules with DNA nanostructures for directing their higher‐order assembly. The hydrophobicity‐driven hybrid DNA nanostructure assembly strategies remain largely unexplored due to various challenges, including the unbiased aggregation of hydrophobic molecules in an aqueous medium, which highlights the importance of using appropriate hydrophobic molecules and experimental conditions to enable a highly programmable hydrophobicity‐based assembly strategy. On the other hand, higher‐order assembly of DNA nanostructures has been well explored by mainly using sticky ends with complementary sequences via Watson–Crick base pairing^{[ 38} , ³⁹ , ⁴⁰ , ⁴¹ , ^{42 ]} and blunt end stacking between helices.^{[ 43} , ⁴⁴ , ^{45 ]} Hence, there is a need to explore alternative strategies to drive the self‐assembly process, which can also provide additional orthogonal interactions beyond hybridization and offer the potential advantage of integrating specific functionality into the assembled hybrid structures.

In this work, we demonstrate a systematic approach to hydrophobicity‐guided higher‐order assembly using carefully designed DNA model systems, ranging from simple DNA tiles to complex DNA origami, functionalized with small hydrophobic molecules as programmable building blocks. We first design and synthesize five different hydrophobic molecules and conjugate them onto a DNA strand. We then employ a T‐shaped DNA tile to incorporate the hydrophobic molecules at the terminal position of its vertical arm to show the hydrophobicity‐driven formation of ladder‐shaped nanostructures. Next, we design a four‐arm DNA tile with hydrophobic modifications at the terminals of its arms to demonstrate the cooperative effect of base pairing and hydrophobic interactions for the formation of 2D arrays. Expanding this strategy, we introduce it to a curved half‐pie DNA origami to compare the efficiency of the five hydrophobic molecules in guiding the higher‐order assembly into defined superstructures. Finally, we use a kite‐shaped DNA origami as a model system to study the effect of the number and spatial distribution of hydrophobic probes on higher‐order assembly.

Results and Discussions

Design and Synthesis of Hydrophobic DNA Probes

Five hydrophobic (HB) molecules, namely HB1, HB2, HB3, HB4, and HB5 with different chemical structures were designed and synthesized. Dendritic molecule HB1 with two C8‐alkyl chains is expected to provide the highest hydrophobicity, while HB2 and HB3 with glycol chains of varying lengths are relatively more hydrophilic. HB4 with a pyrene moiety and HB5 with a unique propeller‐shaped tetraphenylethylene (TPE) core are expected to aggregate via hydrophobic interaction through π–π stacking. When aggregation occurs, the intramolecular free rotation of the aromatic phenyl rings in HB5 will be arrested due to the hydrophobic‐stacking‐induced aggregation, resulting in aggregation‐induced emission (AIE) effect that can be used to characterize such an aggregated state.^{[ 46} , ^{47 ]} The hydrophobic molecules were specifically designed to avoid self‐aggregation after DNA conjugation (e.g., use of oligoethyleneglycol linkers, careful tuning of carbon chain length and branching), ensuring that the molecule‐DNA conjugates remain in a stable single‐stranded DNA form in an aqueous medium. The azide derivatives of these chosen hydrophobic molecules were synthesized by multistep organic syntheses and characterized by spectroscopic methods. All the hydrophobic molecules were conjugated to dibenzocyclooctyne (DBCO)‐modified single‐stranded DNA (DBCO‐TEG‐DNA, TEG = tetraethylene glycol) via strain‐promoted azide‐alkyne cycloaddition (SPAAC) reaction to form the hydrophobic DNA probes (Figure 1a). This resulted in the formation of five DNA amphiphiles, namely, DNA‐HB1, DNA‐HB2, DNA‐HB3, DNA‐HB4, and DNA‐HB5 (Figure 1b). The details of the synthesis, purification, and characterization of DNA hydrophobic conjugates are provided in supporting information (Scheme S1–S5, Figures S1 and S2).

Figure 1.

a) Schematic illustration of strain promoted cycloaddition reaction between DBCO‐TEG DNA and azide derivative of the small hydrophobic molecule. b) Chemical structure of DNA‐HB1, DNA‐HB2, DNA‐HB3, DNA‐HB4, and DNA‐HB5 probes, respectively. Schematic illustration of hydrophobicity driven c) ladder formation by T‐tile assembly, d) 2D array formation by four‐arm junction tile assembly, higher order assembly of e) half‐pie origami with 11 possible probe locations on one side and (f) Kite origami with 14 possible probe locations on each side depicting the most frequently observed configurations after assembly.

Design and Assembly of Hybrid DNA Nanostructures with Hydrophobic Molecules

To demonstrate the capability of the chosen hydrophobic molecules to drive the higher‐order assembly of DNA nanostructures, four different DNA nanostructures with varying geometries, complexity, and molecular weights were designed: (1) a T‐shaped DNA tile (Figure S3, Table S1), (2) a four‐arm DNA tile (Figure S18, Table S11), (3) a half‐pie DNA origami (Figure S25, Table S12), and (4) a kite‐shaped DNA origami (Figure S38, Table S13). The hydrophobic molecules were incorporated into the designed locations of these DNA nanostructures through hybridization between the overhang DNA strands and the hydrophobic‐molecule‐modified DNA, leading to target higher‐order assemblies (Figure 1c–f). The hybrid T‐shaped tile assembly (Figure 1c) and four‐arm tile assembly (Figure 1d) were generated by annealing the hydrophobic probe together with the other DNA strands in a one‐pot assembly method. The unsymmetric half‐pie origami was designed to possess 11 binding sites along the edges of its wider side (Figure 1e). The kite‐shaped origami was designed to have 14 potential binding probes on each of its two longer edges, making it an ideal model system to study the influence of the number and spatial arrangements of hydrophobic modifications to drive efficient and controlled assembly (Figure 1f). In the origami systems, a two‐pot annealing strategy was implemented, wherein the hydrophobic probe was added to a preformed origami and annealed from 45 ^οC to 15 ^οC over 13 h. The annealing program was optimized to promote the modified DNA binding onto DNA origami while maintaining the origami structures intact (annealing protocols are provided in the supporting information).

Controlled Higher‐Order Assembly Mediated by Hydrophobic Interactions

We first used the simplest T‐shaped DNA tile^{[ 48 ]} to incorporate the hydrophobic molecule at the terminal position of its vertical arm, serving as the building block for 1D assembly. The short single‐stranded overhangs of the left horizontal arm, 1 (3 nucleotides) and 2 (6 nucleotides), were designed to be complementary to the unpaired DNA segments on the right horizontal arm, 1′ and 2′, respectively, facilitating Watson–Crick pairing‐based linear assembly in the horizontal direction. Hydrophobic modification at the terminal of the vertical arm enabled the dimerization of two single‐layered linear assemblies to form a ladder‐shaped structure (Figure 2a), as confirmed by atomic force microscopic (AFM) analysis (Figures 2b and S5). Control experiments were performed using the “nonclicked” DBCO‐TEG‐DNA instead of the DNA hydrophobic conjugates, which resulted in the formation of the linear assemblies driven solely by DNA hybridizations (Figures 2b(i) and S4).

Figure 2.

a) Schematic illustration of control T‐Tile assembly depicting single layer formation in the presence of only DBCO‐TEG modification (top) and ladder formation in the presence of hydrophobic molecules (bottom). b) AFM image of (i) control T‐Tile assembly showing only single layer formation using DBCO‐TEG‐DNA in place of HB modified DNA, (ii–vi) ladder formation induced by hydrophobic interactions between linear T‐tile assemblies guided by DNA sticky end interactions along the horizontal direction (scale bar: 200 nm, inset: 50 nm). c) Statistical analysis (from AFM images) comparing the length distribution of formed ladders by the incorporation of hydrophobic molecules (length of single layer assemblies was measured for the control).

To validate the effect of different hydrophobic molecules on the ladder assembly, we analyzed their length distribution by manual measurements of individual assemblies from AFM images (Figures 2c and S6–S14, Tables S3–S10). While the single‐layer control tile assembly grew up to ∼1 µm in length, it was observed that the ladders formed by the hydrophobic interaction could grow up to ∼2 µm with DNA‐HB1, ∼800 nm with DNA‐HB2, ∼700 nm with DNA‐HB3 and, ∼450 nm with DNA‐HB4 and up to 600 nm in the presence of DNA‐HB5. These findings align with our hypothesis that DNA‐HB1, the most hydrophobic conjugate, drives longer‐range assembly, while DNA‐HB2, DNA‐HB3, and DNA‐HB5 lead to moderate assembly lengths, and DNA‐HB4 results in the shortest ladders. The mechanism for ladder formation is proposed to be cooperative, involving DNA sticky‐end interactions along the horizontal arms, which creates a localized, spatially confined hydrophobic environment that attracts another linear T‐Tile assembly, thereby enabling successful dimerization. AFM analysis shows that this ladder formation involves multiple smaller linear assembly fragments coming together, guided by hydrophobic interactions rather than the dimerization of two long, continuous linear assemblies as summarized in Figure S16. DNA‐HB1, being the most hydrophobic among the five conjugates, is found to be able to bring together multiple longer linear assemblies to form extended ladder assemblies.

Next, we extended this hybrid assembly method to 2D array formation driven by the synergistic combination of DNA hybridization in one direction and hydrophobic interactions in the other. To achieve this, we employed a 75^ο angled, four‐arm DNA tile,^{[ 49 ]} programmed with Watson–Crick base pairing interactions for two of the four arms (1–1′ and 2–2′), while incorporating hydrophobic probes at the terminal positions of the other two arms. This design allowed each tile to accommodate two hydrophobic modifications, enabling controlled hydrophobicity‐driven strand assembly into 2D arrays (Figure 3a). To validate the role of hydrophobic interactions, control experiments were performed by incorporating DBCO‐TEG‐DNA in place of the hydrophobic modification. This resulted in the formation of zigzag‐shaped, single‐row assemblies, driven solely by DNA sticky‐end interactions, as shown in the AFM image (Figures 3b and S19). Upon introducing hydrophobic modifications, 2D arrays were formed relying on DNA hybridization in one direction and hydrophobic interaction in the other (Figure 3c). AFM imaging confirmed successful growth of 2D arrays (Figures 3d–h and S20–S24). Among the hydrophobic probes, DNA‐HB1 (Figure 3d) and DNA‐HB5 (Figure 3h) facilitated the formation of the largest arrays, measuring up to 200 nm². The assemblies formed by DNA‐HB2 (Figure 3e), DNA‐HB3 (Figure 3f), and DNA‐HB4 (Figure 3g) were comparatively smaller, likely due to the weaker hydrophobic interactions of these molecules. These 1D and 2D tile assembly results demonstrated a cooperative assembly method of combining base pairing‐induced linear assembly and hydrophobicity‐driven stacking. Furthermore, the AIE properties of the TPE moiety in DNA‐HB5 were used to characterize the 2D array formation. At 2 µM concentration, DNA‐HB5 incorporated into the four‐arm tile assembly showed nearly 2‐fold fluorescence enhancement (λ _ex = 330 nm) compared to its control under the same experimental conditions, strongly indicating the successful stacking of these fluorescent molecules within the 2D arrays (Figure 3i).

Figure 3.

a) Schematic illustration of control four‐arm tile assembly using DBCO‐TEG‐DNA in place of HB modified DNA showing a zigzag shaped single layer assembly due to programmed 1–1′ and 2–2′ DNA sticky end interactions. b) Control AFM images showing single layer formation using DBCO‐TEG‐DNA (scale bar: 400 nm, inset: 100 nm). c) Schematic illustration of 2D array formation by hydrophobic stacking interactions between the linear zigzag assemblies. d)–h) AFM images showing 2D array formation induced by the cooperative effect between hydrophobic interactions and DNA sticky end interactions (scale bar: 400 nm, inset: 100 nm). i) Fluorescence spectral studies on DNA‐HB5 modified tile assembly annealed in 2 µM concentration, showing aggregation induced enhancement in emission intensity by a factor of 1.6 in comparison to 2 µM of DNA‐HB5 alone in 1xTAE buffer, 12.5 mM MgCl₂ at pH 8.2 (Excitation wavelength: 330 nm).

Motivated by the success of tile‐based assembly, we further extended this strategy to DNA origami nanostructures, which possess significantly larger molecular weights and greater structural complexity. To explore hydrophobicity‐driven higher‐order assembly in this system, we designed an asymmetric, curved half‐pie origami with 17 modification sites, including 11 probes on one side and six on the other (not shown, see Supporting Information). The origami was annealed at 2 nM concentration with a five‐fold excess of staple strands, followed by treatment with a 20‐fold excess of the complementary hydrophobic molecule‐modified DNA probes. Initial attempts to incorporate all 17 probes (17‐HB) led to uncontrolled aggregates (Figure S35). The design was revised to incorporate only 11 probes on one side (Figure 4a) to achieve more controlled higher‐order assembly, driven purely by hydrophobic interactions. Control experiments using DBCO‐TEG‐DNA in place of hydrophobic probes showed predominantly monomeric structures, as observed from AFM analysis (Figures 4b and S27). When DNA‐HB1 was introduced into the 11‐probe origami design (11‐HB1), a dramatic decrease in the monomers was observed, with nearly quantitative formation of superstructures, predominantly dimers and trimers (Figures 4c and S28). Interestingly, dimers and trimers consistently dominated the assembled superstructures formed by DNA‐HB1, DNA‐HB2, DNA‐HB3, and DNA‐HB5. AFM (Figures 4c–f, S28–S32, S34) and Agarose gel electrophoresis (Figure 4g) analyses together confirmed the formation of defined higher‐order structures in the form of predominantly dimers, trimers, and tetramers (inset, Figure 4c–f shows the most frequently observed configurations after assembly for each hydrophobic modification). Statistical analysis from manually counting ∼600 individual origami per hydrophobic modification from AFM images (Figure S28) revealed that only 3.7% monomers remained after assembly, with 41.8% dimers and 38.5% trimers in the case of DNA‐HB1, demonstrating a highly efficient higher‐order assembly. DNA‐HB5 exhibited a similar effect, reducing monomers to 3.1%, with 43.5% dimers and 34.4% trimers in the final assembly (Figure S34). DNA‐HB2 (Figure S30) and DNA‐HB3 (Figure S32) also promoted dimers (DNA‐HB2: 24.87%, DNA‐HB3: 21.1%) and trimers (DNA‐HB2: 14.9%, DNA‐HB3: 12.0%), though less efficiently than DNA‐HB1 and DNA‐HB5, likely due to their weaker hydrophobic interactions (Figure 4h). Interestingly, the π–π stacking interactions from DNA‐HB4 were too weak to induce significant higher‐order assembly in this origami system (Figure S33). The observed preference for dimer and trimer assemblies across all these cases may be attributed to the morphological constraints imposed by the origami shape, size, and hydrophobic volume per origami, which likely entropically favor these compact, stable configurations over larger assemblies. Further lowering the number of hydrophobic modifications from 11 to 6 was demonstrated to shift the distribution more to dimers with a corresponding reduction of trimer assemblies (Figure S36). Additionally, we wanted to test the thermal stability of these hydrophobic higher order assemblies in comparison with traditional sticky end hybridization to further motivate this orthogonal approach. For this, we designed half‐pie origami with six possible locations on the right side for sticky end or hydrophobic modification. Impressively, it was observed through Agarose gel electrophoresis and AFM that the higher order assemblies induced by hydrophobic interactions were stable even on heating at 60^οC for 3 h while dimers formed by sticky end hybridization broke down into monomers at 50^οC (Figure S37).

Figure 4.

a) Schematic illustration showing formation of curved half‐pie shaped origami structure with 11 probes on one side. AFM images showing b) control assembly by addition of DBCO‐TEG‐DNA in place of HB modified DNA and higher order assembly of the half‐pie origami guided by c) DNA‐HB1, d) DNA‐HB2, e) DNA‐HB4, and f) DNA‐HB5, respectively (scale bar: 800 nm) with corresponding cropped images showing the most predominant conformations observed after assembly in each case (scale bar: 100 nm). g) 0.8% Agarose gel electrophoresis showing higher order bands in the presence of HB molecules corresponding to origami design with 11 probes on one side (11‐HB). h) Statistical analysis by manual counting from AFM images (∼600 origami per sample) showing the distribution of monomers, dimers, trimers, tetramers, and pentamers achieved by hydrophobic modification with DNA‐HB1, DNA‐HB2, DNA‐HB3, and DNA‐HB5 with a schematic illustration of the predominant conformations observed after assembly.

Finally, we studied a model system in which the numbers and spatial distribution of hydrophobic probes could be precisely controlled to assess their impact on the higher‐order assembly. For this purpose, we designed a kite‐shaped origami with 14 potential modification sites along each long edge for hydrophobic functionalization. Nine variations of the Kite origami were annealed at 2:10 nM scaffold‐to‐staple concentration, with each design incorporating a distinct number and arrangement of hydrophobic probes (Figure 5a). We selected DNA‐HB1 probe to be integrated into the kite origami because this probe was experimentally determined to be highly effective for hydrophobicity‐facilitated assembly of the tiles and half‐pie structure. AFM analyses (Figure 5b) and agarose gel electrophoresis (Figure S48) were used to characterize the higher‐order assembly. Statistical quantification performed through manual counting of AFM images is shown in Figure 5c (Figure S39–S47 for yield counting). Among the first three designs, kite origami‐I with 14 probes on one of its sides and kite origami‐III with seven probes on the bottom half of one side produced the most defined higher‐order assembly (I: 22.6% dimers, 39.9% trimers, and 23% tetramers; III: 35.2% dimers, 27.7% trimers, and 19.9% tetramers). Kite origami‐II, which had seven hydrophobic tags at the central region of one side, had a comparatively higher proportion of monomers (20.6%) with dimers (20.0%), trimers (28.7%), and tetramers (19.7%) again dominating the higher‐order assembly. The origami IV, V, and VI were designed with both sides modified with seven or 14 probes, the AFM images revealed their uncontrolled aggregation. These results aligned well with the half‐pie system, in which the predominantly dimer and trimer formation was generated from the design of hydrophobic probes, which were localized on one side rather than two sides. The assembly results from Kite origami‐I, II, III also demonstrated the impact of the spatial distributions of probes, suggesting the importance of displaying probes near the central vertex. Therefore, kite origami‐VII, VIII, and IX were designed to have 1, 2, and 3 probe pairs, respectively, near the lowermost vertex of the kite origami. Statistical analyses from AFM images showed effective formations of dimers and trimers in designs VIII and IX, with fewer tetramers and higher‐order aggregates, while kite origami‐VII with only one probe on each side showing a significantly higher proportion of monomers (58.0%), which can be explained by the lower hydrophobic volume imparted by the fewer number of hydrophobic tags in design VII.

Figure 5.

a) Schematic illustration showing nine different designs of Kite origami by varying the probe location and number (N = total number of hydrophobic probes per origami design), b) AFM image showing higher order assembly of Kite origami designs I, III, VI, and IX, respectively with DNA‐HB1 modification (scale bar: 800 nm) with corresponding cropped AFM images showing the predominant higher order assembly conformations (scale bar: 100 nm). c) Statistical analysis from AFM images showing the distribution of monomers, dimers, trimers, tetramers, pentamers, and higher order aggregates achieved by DNA‐HB1 hydrophobic modification with a schematic illustration of the predominant conformations observed after assembly.

Conclusion

In this work, we first demonstrated the efficient assembly of DNA tiles to form linear and 2D arrays by the incorporation of hydrophobic and DNA hybridization interactions. This hydrophobicity‐driven assembly strategy was also adapted into origami systems using a curved half‐pie and a kite‐shaped origami nanostructure to demonstrate their higher‐order assembly purely guided by hydrophobic interactions. Among the five synthesized and conjugated hydrophobic probes, DNA‐HB1 exhibited the highest assembly efficiency, suggesting that the hydrophobic alkyl chains are more effective than the glycol chains of DNA‐HB2 and DNA‐HB3 or π–π stacking interactions in DNA‐HB4. The number and spatial distribution of the hydrophobic modifications were also identified as critical parameters in directing well‐defined higher‐order assemblies instead of uncontrolled aggregates, as demonstrated in the Kite origami system. Additionally, DNA‐HB5, featuring AIE properties, showed strong hydrophobic behavior comparable to DNA‐HB1 due to the stacking of its phenyl rings in the aggregated state, which in turn resulted in significant fluorescence enhancement. This unique optical property of DNA‐HB5, when incorporated into DNA‐based nanostructures, holds promise for bioimaging and biosensing applications. All these hydrophobic molecules studied can be easily modified with specific functional groups, which can enable the formation of hybrid nanostructures that have applications as responsive nanomaterials. For example, the aldehyde group in DNA‐HB3 can form reversible linkages such as imine,^{[ 50 ]} hydrazone^{[ 51 ]} and acetals^{[ 52 ]} with complementary functional groups making them potential candidates for various biomedical and material science applications which are currently being explored.

Our strategy advances this field by providing an orthogonal assembly method to better understand and characterize the interplay between hydrophobic and DNA‐based interactions, thereby contributing to a growing library of self‐assembly modules. Despite the versatility of DNA origami higher order assemblies that have already been explored, they are limited in that these aggregates break down at higher temperatures. The high thermal stability of hydrophobicity mediated higher order assembly is, therefore, of great potential for various bottom‐up fabrication methods that require the origami system to be heated to high temperatures. This hydrophobicity‐guided assembly approach, therefore, serves as a valuable addition to traditional DNA self‐assembly strategies, offering both structural programmability and functional versatility even at higher temperatures. Future extensions of this methodology could explore other hydrophobic moieties with specific structural and functional features, guiding the development of dynamic DNA nanodevices. This may be achieved by the incorporation of suitable drug molecules in the hydrophobic region, which can be released only in response to specific stimuli like pH and disease biomarkers at targeted locations. The hydrophobic component of these hybrid systems can also serve as anchors for lipid membrane binding to study various biomimetic processes.

Supporting information

Additional experimental materials and methods, including synthesis of hydrophobic molecules, DNA assembly, AFM imaging, gel electrophoresis, and fluorescence spectroscopy. The file also contains Supporting Figure S1−S48 and Supporting Table S1−S13.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

Supporting information

Saji M., Perumal D., Yang Qi, Jaekle F., Zhang F., Angew. Chem. Int. Ed. 2025, 64, e202507844. 10.1002/anie.202507844

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article.