Design, Synthesis, and Optimization of New Inhalation Carriers: DBBB Series Compounds

Abstract

In the development of dry powder inhaler (DPI) formulations, the choice and optimization of carriers are critical, as they directly impact not only drug stability and bioavailability but also patient adherence. This study sought to create and synthesize a novel DPI excipient with enhanced qualities relative to the current excipient, i.e., fumaryl diketopiperazine (FDKP). In this work, FDKP's framework was utilized to synthesize a variety of novel compounds (DBBB1–15), preserving the diketopiperazine ring and symmetrical branched chains while implementing structural alterations. The artificial intelligence software Schrödinger was employed to screen these chemicals for potential possibilities. As a result, DBBB6 was selected because of its advantageous look and physicochemical properties, including a greater pK_a (reduced acidity) when compared with FDKP. The synthesis method for DBBB6 was refined, resulting in a 9.7% yield. Significantly, investigations involving rats demonstrated that DBBB6 did not induce coughing, a possible adverse effect associated with FDKP. The results indicate that DBBB6 is a viable alternative to FDKP as a DPI excipient. Its improved tolerability profile suggests a potential for reduced adverse effects. Additional studies are required to comprehensively assess its safety and efficacy for clinical application.

Keywords

DBBB series compounds - fumaryl diketopiperazine - dry powder inhaler

Introduction

Dry powder inhalation (DPI) is a well-established administration approach for drug delivery and is widely used to treat respiratory diseases such as asthma and chronic obstructive pulmonary disease.[1] In the development of DPI formulations, the choice and optimization of carriers are critical, as they not only directly impact drug stability and bioavailability but also patient adherence. Currently, lactose is the most commonly used carrier in DPI formulations owing to advantages including ease of access, good safety, and broad compatibility with most active pharmaceutical ingredients.[2] [3] However, as a reducing sugar, β-anhydrous lactose is hygroscopic and can react with primary and secondary amines when stored under conditions of elevated temperature and high humidity.[4] Hence, lactose is not suitable as a carrier for primary amines, peptides, and protein drugs. Besides, micronized trehalose,[5] glucose,[6] mannitol,[7] and cyclodextrin[8] [9] are also used as carriers in DPI formulations. However, DPIs prepared with these excipients as raw materials may have problems such as excessively large particle size and poor lung deposition.[10] In addition, DPI performances are affected by the shape, size, density, moisture, and surface properties of the particles,[3] and there may be interactions between the carriers and the drug, resulting in decreased drug stability or reduced bioavailability. Therefore, it is of great significance to explore new carriers for DPIs. Current methods employ nanoparticles and biocompatible materials to improve drug solubility for inhalation, or use chemical and physical modification—such as coating technology—to enhance powder fluidity, reduce agglomeration, and ultimately improve the atomization performance of dry powders. Unfortunately, these approaches are difficult to translate into clinical practice.

As one of the most successful DPI products, Afrezza, the dry powder inhaled insulin developed by MannKind, was approved by the U.S. Food and Drug Administration (FDA) on June 27, 2014, and was launched in the U.S. market in 2015, providing a new direction for the research and development of peptide drugs. In the formulation of Afrezza, fumaryl diketopiperazine (FDKP), with a chemical structure ([Fig. 1]) of (2E,2'E)-4,4'-(((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(4-oxobut-2-enoic acid), is used as the main carrier, which is a type of pulmonary inhalation excipient recently approved by the FDA. Under a weakly acidic environment, FDKP self-assembles into microspheres[11] ([Fig. 2]), which have a large specific surface area.[12] The method of drug loading using FDKP is referred to as Technosphere technology. Compared with injectable insulin, Afrezza can better simulate insulin secretion in healthy humans, improve postprandial blood sugar control, and significantly reduce the risk of hypoglycemia and weight gain. However, the application of Afrezza may cause adverse respiratory reactions such as cough and bronchospasm, leading to poor patient compliance.[13] Studies found that this side effect is attributed to the two acidic carboxyl groups at the end of the FDKP branch chain.[14] Due to the aberrant bronchopulmonary C fibers and Aδ afferents, the inhalation of acidic substances can cause coughing.[13] [15] [16] [17] Therefore, it is meaningful to develop new excipients that can reduce such side effects.

Fig. 1 Chemical structure of FDKP.

Fig. 2 Representative image of FDKP self-assembled microspheres.

Based on the core structure of FDKP, we aimed to design and synthesize a new series of compounds (DBBB1–15) as inhalation carriers. To obtain compounds that surpass the physical morphology and properties of FDKP microspheres, while reducing the associated side effects in patients, we designed a compound library. This was then screened using the leading artificial intelligence (AI) drug discovery software, Schrödinger, and the preferred compounds were finally obtained. To improve the yield of the synthesis, a high-throughput synthesis route was designed, and a series of compounds was successfully prepared using this method. These compounds were subjected to further screening according to their physicochemical properties, and as a result, DBBB6 with a honeycomb structure and a pK_a greater than that of FDKP was obtained. Finally, the DBBB6 synthesis route was optimized to establish a scalable process.

Results and Discussion

Characterization of DBBB Series Compounds

The synthesis route of DBBB compounds was shown in [Fig. 3]. The structures of BBB1–15 were validated with ¹H NMR, ¹³C NMR, FTIR spectra, and ESI-MS data, and the spectra were shown in the [Supporting Information] (available in the online version).

Fig. 3 Synthesis route for DBBB series compounds.

Appearance and Morphology Inspection

The appearance and morphology of the obtained DBBB series compounds are shown in [Fig. 4]. DBBB3 and DBBB9 were eliminated from consideration, because the compounds were in the form of a honey or oil, which is not suitable for use as a carrier in DPI.

Fig. 4 Appearance and morphology of DBBB series compounds.

Next, the morphology of other DBBB series compounds was observed by optical microscopy, and the images are shown in [Fig. 5]. During the observation, some samples, such as DBBB1, DBBB5, DBBB7, and DBBB14, showed softening phenomena, implying that there are certain problems with their stability, which are not suitable for use as carriers for DPI formulations.

Fig. 5 Optical microscopic images of DBBB series compounds.

After screening by optical microscope, the remaining samples were observed further using a scanning electron microscope (SEM), and the resulting photos are shown in [Fig. 6]. The SEM results revealed that, among these compounds, only the DBBB6 microspheres displayed a honeycomb structure on their surface. DBBB6 was also observed from different angles, and the SEM images are shown in [Fig. 7]. This structure resulted in a large number of irregular voids on the microsphere surface, giving it a large surface area. Therefore, DBBB6 is the most promising candidate.

Fig. 6 SEM photos of DBBB series compounds. SEM, scanning electron microscope.

Fig. 7 Representative SEM images of DBBB6 with scale bars indicating (A) 10 μm and (B) 200 μm, respectively.

For most DPIs that use excipients, these substances often account for a large proportion of the formulation, indicating that the excipients play a key role in the safety of DPI products. If highly toxic substances are introduced during the synthesis of excipients, the subsequent applications may pose increased safety risks. In the synthesis process of FDKP, highly toxic p-nitrobenzene derivatives are used as starting materials ([Fig. 8]).[18] If these toxic substances cannot be completely removed, there will be great safety risks. However, the synthesis route of the DBBB series compounds designed in this study eliminates the nitrobenzene structure. It uses the general synthesis method of condensing monomethyl (or ethyl) esters of monocarboxylic acids. This method efficiently prepares a series of compounds and greatly reduces the risk of toxic raw materials being present in the final product.

Fig. 8 Original synthesis route of FDKP.

pK_a Determination

The pK_a values of DBBB6 and FDKP predicted by ACD/Percepta Portal software (ACD/Labs, Inc., Canada) are shown in [Figs. 9] and [10], respectively. The pK_a value of DBBB6 was 4.4 ± 0.4, and the pK_a value of FDKP was 3.6 ± 0.8. Obviously, the acidity of DBBB6 was weaker than that of FDKP.

Fig. 9 pK_a value of DBBB6 predicted by ACD/percepta portal.

Fig. 10 pK_a value of FDKP predicted by ACD/percepta portal.

The pK_a values of DBBB6 and FDKP were further predicted by potentiometric titration. The result showed that the pK_a value of DBBB6 is 3.56 ± 0.47, and the pK_a value of FDKP is 3.30 ± 0.60. Although the result determined by potentiometric titration was slightly lower than that predicted by ACD/Percepta Portal (ACD/Labs, Inc., Canada), the trends of the two methods remained the same; that is, DBBB6 is weaker in acidity than FDKP.

Optimization of the DBBB6 Synthesis Process

Although a high-throughput synthesis method was used, the preparation scale of DBBB6 was limited to the milligram level. To increase its synthesis scale and meet the needs of subsequent research, it is necessary to optimize its synthesis parameters. The reaction solvent for the first step, the temperature and hydrogen pressure for the second step, and the equiv. ratio of regents for the third step (IM-2: 3-(methoxycarbonyl)bicyclo[1.1.1]pentane-1-carboxylic acid (IM-3–6): 2-(7-azabenzotriazol-1-yl)-N,N,N',N'-tetramethyluronium hexafluorophosphate [HATU]), has been assessed, respectively.

Reaction Solvent Screening

Compared with m-cresol and diphenyl ether, the product obtained using N-methylpyrrolidone as the reaction solvent had the best morphology and the highest yield, which reached 48.3 to 55.2% ([Table 1]).

Table 1 Results of using different reaction solvents

Solvent	Reaction conditions	Results
m-Cresol	P2O5, 200°C	The posttreatment was highly exothermic, and basically no product was generated
Diphenyl ether	200°C	The properties of the product are not good, with a 10% yield
N-Methylpyrrolidone	160°C	48.3–55.2% yield

Temperature and Hydrogen Pressure Screening

To investigate the effect of temperature, the hydrogen pressure was fixed at 10 psi, and the reactions were performed at 25, 35, and 50°C, respectively. The results showed that the yields were low at all temperatures but appeared relatively higher at higher temperatures. To improve the conversion rate, the temperature was fixed at 50°C and the hydrogen pressure was increased to 100 psi. The conversion rate reached 100%, indicating that a higher hydrogen pressure is beneficial to the reaction ([Table 2]).

Table 2 Results of using different temperatures and hydrogen pressures

Temperature (°C)	Hydrogen pressure (psi)	Results
25	10	Yield: 0
35	10	Yield: low
50	10	Long reaction time, slow conversion, and low yield
50	100	Conversion rate: 100%

Equiv. Ratio of IM-2/IM-3-6/HATU

Three equiv. ratios of IM-2/IM-3-6/HATU were investigated in this study. With the increase of side chain and HATU input, the conversion rate gradually increased, and the yield of the product also increased continuously. Therefore, 1:4:4 is proven to be the optimal equivalent ratio.

The synthesis of DBBB6 involves multiple steps. The preferred solvent for the first step is N-methylpyrrolidone, and the preferred reaction temperature is 160°C. The selected reaction temperature for the second step is 50°C, and the hydrogen pressure is 100 psi. The preferred stoichiometric ratio of IM-2/IM-3-6/HATU for the third step is 1:4:4. After optimizing these synthesis steps and calculating the yield for each step, the yield for DBBB6 reached 9.7%, which met the requirements for subsequent research.

Animal Study

Quantitative analysis revealed distinct cough response patterns among the treatment groups. The DBBB6-treated cohort demonstrated 2.1 ± 0.8 coughs/30 min, showing no statistical difference from saline controls (1.8 ± 0.6 coughs/30 min; p > 0.05). Notably, animals exposed to FDKP exhibited a significantly higher cough frequency (6.5 ± 1.2 coughs/30 min), representing a 3.6-fold increase versus the DBBB6 group and a 3.2-fold increase versus saline controls (both p < 0.01). Pulmonary function testing showed transient airway hyperresponsiveness in the FDKP group (∼15% increase), whereas DBBB6 administration maintained stable respiratory mechanics, suggesting that DBBB6 compensates for the FDKP-induced cough defect.

Compared with lactose, FDKP is a newly approved carrier that was first used in Afrezza, but research into its application is insufficient. According to reports, 27.8% (257/923) of patients in the Afrezza group experienced coughing, and poor patient compliance also led to its unsatisfactory sales in the United States.[19] The coughing side effect is attributed to the two acidic carboxyl groups at the end of the FDKP side chain. Inhaling acidic substances can trigger coughing via aberrant bronchopulmonary C fibers and Aδ afferents.[15] DBBB6 that retains the diketopiperazine parent ring structure of FDKP but modifies its side chains. This results in an improved pK_a value and reduced acidity, thereby reducing the risk of coughing in patients. This effect has been confirmed in animal experiments using SD rats as a model. Moreover, pulmonary hydrolase activity is low at high pH,[20] thus reducing the acidity of DBBB6 also improves the stability of peptide drugs in the lungs.

Conclusion

In this study, we designed and synthesized a series of DBBB compounds by modifying the side chains of FDKP. Morphological analysis identified DBBB6 as the most promising candidate for DPI formulations. We subsequently optimized the synthesis yield of DBBB6 by adjusting the key reaction parameters, including solvent, temperature, hydrogen pressure, and feed ratio. Animal studies demonstrated that DBBB6's reduced acidity prevents it from inducing cough, a significant advantage over FDKP. Together, these findings suggest DBBB6 is a superior and potentially more effective excipient in DPI formulations.

Experimental Section

Materials

All the reagents are commercially available. Phosphorus pentoxide, palladium on carbon, N,N'-diisopropylcarbodiimide, (1R,6R)-6-(methoxycarbonyl) cyclohex-3-ene-1-carboxylic acid, and cis-2-carbomethoxycyclohexane-1-carboxylic acid were acquired from Sigma-Aldrich. HOBt, m-cresol, DIPEA, monomethyl adipate, monomethyl succinate, monomethyl malonate, monomethyl 5-heptene-2,3-dicarboxylate, bicyclo(1.1.1)pentane-1,3-dicarboxylic acid, 1-methyl ester, monomethyl 1,2-cyclopropanedicarboxylate, monomethyl 1,1-cyclopropyldicarboxylate, monoethyl cyclobutyl-1,1-dicarboxylate, monomethyl 2,2-dimethylmalonate, 1-methyl 2,2-dimethylsuccinate, (1S,6R)-6-(methoxycarbonyl)cyclohex-3-ene-1-carboxylic acid, monomethyl methylglutarate, L-lysine, N6-Cbz-L-lysine, glacial acetic acid, monomethyl fumarate, 5-chloro-2-pentanone, N-benzylethanolamine, terephthalic acid, isophthalic acid, phthalic acid, and N-methylpyrrolidone were acquired from Anage Chemical. Monomethyl itaconate and monomethyl 2,6-pyridinedicarboxylate were acquired from Adamas, whereas semaglutide (SMG) was sourced from Yangzhou Aoruit Pharmaceutical Co., Ltd. Hydrochloric acid and ammonia were acquired from Sinopharm Chemical Reagent Co., Ltd. All reagents were of chemical or analytical grade and utilized as received.

Structural Design of DBBB Series Compounds

To build prospective molecules with enhanced physicochemical properties, the diketopiperazine parent ring in FDKP was preserved while the branching chain structure was modified. All molecules in the compound library ([Supporting Information], available in the online version) were novel entities, provisionally designated as the DBBB series based on their general formula ([Fig. 3]). Given that molecules typically exist in a low-energy state under natural conditions, and considering that the microsphere formation process may engage hydrogen bonds, van der Waals forces, and other interactions, the drug discovery AI software Schrödinger (Schrödinger, Inc., New York, United States) was employed to preserve the diketopiperazine parent ring structure and the symmetry of the two side chains, while adjusting the distance between the diketopiperazine carbonyl group and the terminal carboxyl and amide groups of the branch chain to mitigate the acidity of the novel compound. The chemical library was molecularly screened utilizing FDKP as a reference[21] to identify compounds exhibiting specific solid shape, substantial surface area, and drug loading capacity.

Prepartion of a Series of DBBB Compounds

Synthesis of IM-1

The IM-1 was synthesized by a dehydration cyclisation process. In summary, 30 g of ε-benzyloxycarbonyl-L-lysine (SM-1), 50 g of m-cresol, and 5 g of phosphorus pentoxide were introduced into a conical flask and heated to 200°C. Throughout the reaction, water was distilled, and subsequently, the reactants were cooled with a mixed solution of water and sodium hydroxide (10:1) to generate a precipitate. The precipitate was isolated and washed with 30 mL of ethanol, followed by filtration to provide 20.6 g of the crude product, 2,5-bis[4-(N-benzyloxycarbonyl)aminobutyl]-3,6-diketopiperazine (SM-2). The crude product was heated to 100°C in 100 mL of glacial acetic acid solution, followed by the addition of 30 mL of pure water for cooling. The crystals were ultimately washed with 50 mL of glacial acetic acid solution, yielding 11.2 g of the purified IM-1.

Synthesis of IM-2

A hydrogenation process was conducted to synthesize IM-2. A total of 11.2 g of refined IM-1 was dissolved in 50 mL of glacial acetic acid solution, and a catalyst (10% palladium on carbon) was introduced to facilitate a hydrogenation reaction in a hydrogen-filled reactor. The resultant combined solution was subjected to cooling and filtration. Thereafter, the filtrate was distilled to eliminate the glacial acetic acid, resulting in the acquisition of the acetate of 2,5-bis(4-aminobutyl)-3,6-diketopiperazine (IM-2).

Synthesis of IM-4

A condensation process was performed to synthesize IM-4. The IM-2 product was dissolved in DMF. IM-3, triethylamine, and HATU were introduced to facilitate the condensation reaction at ambient temperature. Following the detection of the liquid phase, indicating the disappearance of the raw material, 100 mL of water was introduced, and the resultant solid material was isolated, yielding 5.8 g of a crude solid product of compound IM-3. Subsequently, 20 mL of glacial acetic acid was incorporated into the synthesized compound IM-3, followed by the addition of 20 mL of water for cooling purposes. Subsequently, the crystals were rinsed with 10 mL of glacial acetic acid solution, yielding the pure compound IM-4.

Synthesis of a Series of DBBB Compounds

A series of DBBB compounds was produced via a saponification reaction. A total of 20 mL of a mixed solution of methanol and sodium hydroxide was added to the product generated in the preceding stage and heated to approximately 70°C. After filtering, glacial acetic acid was added, and the mixture was cooled to room temperature. The solid was then isolated and washed with water to get a crude product (the target DBBB compound). To purify the product, trifluoroacetic acid was incorporated into the crude product and heated to 90°C. The combination was subsequently filtered, and the filtrate was further chilled by the addition of glacial acetic acid. Subsequently, the solid material was isolated, washed with methanol and filtered water, and refined by column chromatography, yielding the target compound.

Characterization of DBBB Series Compounds by Nuclear Magnetic Resonance Hydrogen Spectrum (¹H NMR), Carbon Spectrum (¹³C NMR), Fourier-Transform Infrared Spectroscopy, and Electrospray Ionization Mass Spectrometry

The structures of DBBB series compounds were confirmed through the measurement of ¹H NMR (Bruker, Rheinstetten, Germany) and ¹³C NMR (Bruker, Rheinstetten, Germany) of the intermediate products and final DBBB compounds, utilizing deuterated dimethyl sulfoxide (DMSO) or deuterated methanol (MeOD) as solvents and tetramethylsilane (TMS) as the internal standard. Chemical shift (δ) is expressed in units of parts per million (ppm). Multiplicities are abbreviated as singlet (s), doublet (d), triplet (t), quartet (q), and multiplet (m). Coupling constants (J) are reported in Hertz (Hz).

The product combination and KBr were pulverized in an agate mortar, transferred to a tablet pressing mold, and compressed into tablets, which were subsequently analyzed using an infrared spectrometer (IR Tracer-100, Shimadzu, Japan). To acquire the mass distribution of the product, a suitable quantity of the test material was dissolved in a solvent system comprising trifluoroacetic acid and acetonitrile, and subsequently injected into the ion source. The test parameters included a scanning range of m/z 50 to m/z 900, a nebulization pressure of 30 psi, a drying gas temperature of 350°C, and a spray voltage of 3,000 V.

Characterization of FDKP. Chemically named “(2E,2'E)-4,4'-(((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(4-oxobut-2-enoic acid).” ¹H NMR (400 MHz, DMSO-d ₆) δ 8.44 (m, 2H), 8.12 (m, 2H), 6.85 (d, J = 15.6 Hz, 2H), 6.53 (d, J = 15.2 Hz, 2H), 3.83–3.78 (m, 2H), 3.17- 3.12 (m, 4H), 1.77–1.64 (m, 4H), 1.43–1.26 (m, 8H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 168.07, 167.84, 166.53, 163.00, 136.97, 129.57, 53.88, 53.79, 32.75, 32.02, 28.62, 28.54, 21.80, 21.33. ESI-MS (m/z) calcd. for C₂₀H₂₉N₄O₈ ⁺ [M + H]⁺ 453.19, found: 453.20.

Characterization of DBBB1. Chemically named “3,3′-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))bis(but-3-enoic acid).” ¹H NMR (400 MHz, DMSO-d ₆) δ 8.63 (d, J = 6.0 Hz, 1H), 8.15 (s, 1H), 7.03 (d, J = 15.6 Hz, 1H), 6.60 (d, J = 15.2 Hz, 1H), 3.83 (m, 1H), 3.75 (s, 3H), 3.15 (q, J = 6, 6.4 Hz, 2H), 1.68 (m, 2H), 1.46–1.24 (m, 4H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 176.14, 174.54, 174.11, 173.44, 168.17, 167.93, 125.73, 125.52, 125.10, 124.98, 53.85, 53.73, 32.92, 31.85, 28.71, 28.52, 28.07, 26.64, 26.46, 21.64, 21.03. ESI-MS (m/z) calcd. for C₂₂H₃₃N₄O₈ ⁺ [M + H]⁺ 481.22, found: 481.22.

Characterization of DBBB2. Chemically named “3,3′-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))bis(bicyclo[2.2.1]hept-5-ene-2-carboxylic acid).” ¹H NMR (400 MHz, DMSO-d ₆) δ 12.07 (br, 2H), 8.10–8.09 (m, 2H), 7.96–7.95 (m, 2H), 6.26–6.25 (m, 2H), 6.08–6.06 (m, 2H), 3.80–3.75 (m, 2H), 3.25–3.24 (m, 2H), 3.12–3.10 (m, 4H), 3.10–3.01 (m, 2H), 3.00–2.82 (m, 2H), 2.37–2.35 (m, 2H), 1.71–1.65 (m, 6H), 1.41–1.22 (m, 10H),. ¹³C NMR (100 MHz, DMSO-d ₆) δ 174.69, 172.72, 168.04, 167.86, 163.09, 137.67, 135.21, 53.90, 53.84, 48.55, 47.37, 46.87, 46.32, 44.68, 32.78, 32.10, 28.99, 28.83, 21.78, 21.31, 21.25. ESI-MS (m/z) calcd. for C₃₀H₄₁N₄O₈ ⁺ [M + H]⁺ 585.2919, found: 585.27.

Characterization of DBBB3. Chemically named “5,5′-(((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(3-methyl-5-oxopentanoic acid).” ¹H NMR (400 MHz, DMSO-d ₆) δ 12.14 (s br, 2H), 8.16 (s, 2H), 8.02–8.01 (m, 2H), 3.79–3.78 (m, 2H), 3.12–2.99 (m, 4H), 2.26–2.23 (m, 4H), 2.09–3.07 (m, 6H), 1.98–1.86 (m, 4H), 1.37–1.33 (m, 8H), 1.00 (s, 6H). ¹³C NMR (100 MHz, DMSO-d₆ ) δ 172.77, 172.74, 169.79, 166.85, 166.58, 116.85, 52.58, 52.46, 51.98, 51.73, 43.83, 31.55, 30.63, 27.57, 27.42, 27.29, 23.30, 20.27, 19.72, 16.94, 14.20, 10.66. ESI-MS (m/z) calcd. for C₂₄H₄₁N₄O₈ ⁺ [M + H]⁺ 513.2919, found: 513.28.

Characterization of DBBB4. Chemically named “4,4'-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))dibenzoic acid.” ¹H NMR (400 MHz, DMSO-d ₆) δ 8.86–8.61 (m, 2H), 8.11–8.10 (m, 2H), 7.98–7.97 (m, 4H), 7.57–7.47 (m, 4H), 3.82–3.79 (m, 2H), 3.31–3.21 (m, 4H), 1.75–1.71 (m, 4H), 1.64–1.61 (m, 4H), 1.33–1.30 (m, 4H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 174.86, 172.87, 172.33, 168.74, 168.56, 52.37, 49.40, 28.29, 23.22, 22.59, 21.23. ESI-MS (m/z) calcd. for C₂₈H₃₃N₄O₈ ⁺ [M + H]⁺ 553.2293, found: 553.23.

Characterization of DBBB5. Chemically named “4,4'-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))dibenzoic acid.” ¹H NMR (400 MHz, CDCl₃) δ 8.21–8.15 (m, 2H), 7.76–7.74 (m, 2H), 3.79–3.77 (m, 2H), 3.04–3.02 (m, 4H), 2.38–2.33 (m, 8H), 1.73–1.65 (m, 8H), 1.41–1.19 (m, 8H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 173.80, 173.19, 172.97, 172.72, 172.27, 171.09, 171.02, 169.36, 169.28, 168.73, 168.59, 168.06, 66.98, 63.14, 59.84, 53.78, 51.93, 51.55, 51.30, 45.45, 44.98, 38.39, 38.15, 30.26, 29.91, 29.69, 28.93, 28.77, 28.65, 28.54, 28.36, 27.48, 27.38, 25.57, 25.09, 22.49, 21.24, 21.06. ESI-MS (m/z) calcd. for C₂₄H₃₇N₄O₈ ⁺ [M + H]⁺ 509.2606, found: 509.24.

Characterization of DBBB6. Chemically named “3,3′-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))bis(bicyclo[1.1.1]pentane-1-carboxylic acid).” ¹H NMR (400 MHz, CDCl₃) δ 12.42 (s, br, 2H), 8.07 (s, 2H), 7.78–7.75 (m, 2H), 3.78–3.77 (m, 2H), 3.01–2.97 (m, 4H), 2.12–2.07 (d, 12H), 1.68–1.65 (m, 4H), 1.64–1.26 (m, 8H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 170.88, 168.42, 168.16, 168.05, 167.81, 53.82, 51.68, 51.48, 36.52, 32.71, 32.04, 28.83, 21.68, 21.19. ESI-MS (m/z) calcd. for C₂₆H₃₇N₄O₈ ⁺ [M + H]⁺ 533.2606, found: 533.25.

Characterization of DBBB7. Chemically named “3,3′-(((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(2,2-dimethyl-3-oxopropanoic acid).” ¹H NMR (400 MHz, CDCl₃) δ 4.41–4.34 (m, 1H), 3.98–3.93 (m, 1H), 3.25–3.21 (m, 4H), 1.94–1.75 (m, 4H), 1.58–1.29 (m, 20H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 174.86, 172.87, 172.33, 168.74, 168.56, 52.37, 49.40, 28.29, 23.22, 22.59, 21.23. ESI-MS (m/z) calcd. for C₂₂H₃₇N₄O₈ ⁺ [M + H]⁺ 485.2606, found: 485.24.

Characterization of DBBB8. Chemically named “(1R,1'R,6R,6'R)-6,6'-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))bis(cyclohex-3-ene-1-carboxylic acid).” ¹H NMR (400 MHz, CDCl₃) δ 12.36 (s, br, 2H), 8.28–8.25 (m, 2H), 8.14–8.06 (m, 2H), 5.63–5.56 (d, 2H), 3.77–3.74 (m, 2H), 2.50–1.92 (m, 8H), 1.65–1.64 (m, 4H), 1.34–1.20 (m, 8H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 173.34, 172.36, 171.55, 170.17, 169.48, 168.34, 168.11, 167.85, 53.86, 53.75, 51.81, 51.34, 32.83, 31.92. 28.67, 23.15, 22.91, 22.41, 21.72, 21.24, 20.68, 20.46, 20.07, 13.36, 11.17, 10.08. ESI-MS (m/z) calcd. for C₂₈H₄₁N₄O₈ ⁺ [M + H]⁺ 561.2919, found: 561.26.

Characterization of DBBB9. Chemically named “(1R,2S)-2-((4-(5-(4-((1R,2S)-2-carboxycyclohexane-1-carboxamido)butyl)-3,6-dioxopiperazin-2-yl)butyl)carbamoyl)cyclohexane-1-carboxylic acid.” ¹H NMR (100 MHz, DMSO-d₆ ) δ 8.66–8.59 (m, 2H), 8.38–8.37 (m, 2H), 8.23–8.21 (m, 1H), 3.62–3.51 (m, 14H), 3.10–2.89 (m, 4H), 2.16–1.94 (m, 2H), 1.93–1.75 (m, 2H), 1.75–1.54 (m, 4H), 1.49–1.18 (m, 8H), 1.18–1.03 (m, 2H). ESI-MS (m/z) calcd. for C₂₈H₄₅N₄O₈ ⁺ [M + H]⁺ 565.3232, found: 565.31.

Characterization of DBBB10. Chemically named “1,1'-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))bis(cyclopropane-1-carboxylic acid).” ¹H NMR (400 MHz, DMSO-d ₆) δ 12.19 (brs, 2H), 8.36–8.31 (m, 4H), 3.91–3.74 (m, 12H), 2.97–2.93 (m, 4H), 2.22–2.17 (m, 4H), 2.09–2.01 (m, 8H), 1.65–1.61 (m, 4H), 1.31–1.27 (m, 8H), 0.90–0.71 (m, 6H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 173.41, 172.49, 171.12, 171.10, 168.13, 167.83, 53.84, 53.70, 51.20, 42.01, 32.90, 31.84, 28.77, 28.62, 27.45, 27.34, 21.73, 21.13, 19.31. ESI-MS (m/z) calcd. for C₂₂H₃₃N₄O₈ ⁺ [M + H]⁺ 481.2293, found: 481.20.

Characterization of DBBB11. Chemically named “2,2'-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))dibenzoic acid.” ¹H NMR (400 MHz, CDCl₃) δ 12.85 (s, br, 2H), 8.45–8.44 (m, 2H), 8.09–8.08 (m, 2H), 6.91–6.87 (m, 2H), 6.49–6.46 (m, 2H), 3.80–3.77 (m, 2H), 3.24–3.09 (m, 4H), 1.73–1.71 (m, 4H), 1.42–1.25 (m, 8H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 167.97, 167.70, 165.54, 162.63, 162.54, 137.72, 127.74, 53.78, 53.69, 51.83, 48.13, 47.92, 47.70, 47.49, 47.28, 47.07, 46.86, 32.71, 31.98, 28.56, 28.49, 21.73, 21.27. ESI-MS (m/z) calcd. for C₂₈H₃₃N₄O₈ ⁺ [M + H]⁺ 553.2293, found: 553.21.

Characterization of DBBB12. Chemically named “3,3′-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))dibenzoic acid.” ¹H NMR (400 MHz, CDCl₃) δ 8.43–8.42 (m, 2H), 8.41–8.29 (m, 2H), 8.02–8.01 (m, 2H), 7.58–7.54 (m, 2H), 3.81–3.79 (m, 2H), 3.25–3.22 (m, 4H), 1.75–1.71 (m, 4H), 1.49–1.47 (m, 4H), 1.31–1.23 (m, 4H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 168.12, 167.90, 167.02, 165.34, 134.93, 131.63, 131.37, 131.24, 128.56, 127.94, 53.93, 53.87, 32.78, 32.13, 28.88, 28.80, 21.86, 21.43. ESI-MS (m/z) calcd. for C₂₈H₃₃N₄O₈ ⁺ [M + H]⁺ 553.2293, found: 553.20.

Characterization of DBBB13. Chemically named “4,4'-(((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(3,3-dimethyl-4-oxobutanoic acid)” ¹H NMR (400 MHz, DMSO-d₆ ) δ 4.87 (m, 2H), 4.04–4.02 (m, 4H), 3.98 (s, 1H), 3.68–3.64 (m, 2H), 3.35–3.16 (m, 4H), 2.59–2.47 (m, 4H), 1.83–1.82 (m, 4H), 1.54–1.22 (m, 13H).¹³C NMR (100 MHz, DMSO-d ₆) δ 170.33, 168.14, 55.84, 53.73, 44.74, 38.16, 32.85, 31.89, 28.74, 28.60, 26.34, 25.62, 25.31, 21.69, 21.14. ESI-MS (m/z) calcd. for C₂₄H₄₀N₄O₈ ⁺ [M + H]⁺ 513.6040, found: 513.60.

Characterization of DBBB14. Chemically named “2,2'-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))bis(cyclopropane-1-carboxylic acid).” ¹H NMR (400 MHz, CDCl₃) δ 12.48 (s, br, 2H), 8.64–8.61 (m, 2H), 8.58–8.16 (m, 2H), 3.77–3.66 (m, 2H), 3.12–3.04 (m, 2H), 2.99–2.92 (m, 4H), 2.09–2.07 (m, 2H), 1.74–1.62 (m, 4H), 1.34–1.29 (m, 8H), 1.11–1.06 (m, 4H) ¹³C NMR (100 MHz, DMSO-d ₆) δ 176.26, 175.11, 171.71, 170.66, 170.39, 170.03, 169.80, 169.30, 169.21, 168.11, 129.45, 53.79, 53.69, 51.41. 49.57, 49.42, 48.53, 48.38, 37.83, 36.46, 36.28, 34.91, 31.97, 29.87, 28.69, 28.42, 28.33, 23.37, 22.57, 22.34, 22.04, 21.13, 13.86, 10.87. ESI-MS (m/z) calcd. for C₂₂H₃₃N₄O₈ ⁺ [M + H]⁺ 481.2293, found: 481.20.

Characterization of DBBB15. Chemically named “6,6'-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))dipicolinic acid.” ¹H NMR (400 MHz, CDCl₃) δ 8.35–8.21 (m, 4H), 8.18–8.17 (m, 2H), 4.06–4.00 (m, 2H), 3.65–3.46 (m, 4H), 1.88–1.87 (m, 4H), 1.75–1.70 (m, 4H), 1.51–1.24 (m, 4H). ¹³C NMR (100 MHz, DMSO-d ₆) δ 168.09, 167.72, 165.58, 165.35, 162.99, 162.95, 149.35, 147.92, 147.41, 139.45, 139.32, 126.32, 124.66, 124.45, 53.94, 53.86, 53.04, 41.45, 32.64, 32.07, 29.07, 28.94, 21.70, 21.42, 17.57, 12.55. ESI-MS (m/z) calcd. for C₂₆H₃₁N₆O₈ ⁺ [M + H]⁺ 555.2198, found: 555.20.

Characterization of DBBB Series Compounds by Appearance and Morphology Inspection

Potential candidate chemicals were identified using visual inspection and optical microscopy (Olympus, Shinjuku, Japan). Subsequently, the aesthetically pleasing compounds were subjected to further examination utilizing scanning electron microscopy (Phenom XL G2, Netherlands).

Determination of pK_a of DBBB Series Compounds

The pK_a of DBBB6 was initially calculated using ACD/Percepta Portal software (ACD/Labs, Inc., Canada) and subsequently measured using potentiometric titration. In brief, 0.56 g of potassium chloride was dissolved in 20 mL of deionized water, followed by the addition of 10 mL of acetonitrile and 10 mL of dioxane. The co-solvent was subsequently adjusted to a volume of 50 mL using MeOH. 1 mg of the sample was measured into a vial, and 1.5 mL of co-solvent was automatically dispensed into the vial by the pK_a detector (Sartorius, Göttingen, Lower Saxony, Germany). The apparatus autonomously preacidified the sample solution with 0.5 mmol/L HCl and thereafter conducted three titrations from low to high pH to ascertain the pK_a value. The titrations were performed using the Titrant, 0.1 mol/L NaOH (standardized, carbonate-free) for base addition; and the indicator, phenolphthalein (pH 8.2–10.0, colorless to pink) for basic transitions, and bromothymol blue (pH 6.0–7.6, yellow to blue) for near-neutral pK_a.

Optimization of the DBBB6 Synthesis Process

Effect of Reaction Solvent

To synthesize IM-1, ε-benzyloxycarbonyl-L-lysine (SM-1) was used as the starting material and was added to diphenyl ether and heated to 200°C. After the reaction was completed, sodium hydroxide aqueous solution was added, and the mixture was filtered after cooling, and the filter cake was subsequently washed with ethanol. The obtained product was a yellow-brown solid with a strong odor of the reaction solvent benzocresol. In addition to diphenyl ether, m-cresol and N-methylpyrrolidone were also used as reaction solvents. To determine which is the most suitable solvent, the purity and yield of IM-1 were determined and used as evaluation indicators.

Effect of Temperature and Hydrogen Pressure

To synthesize IM-2, IM-1 was added into a single-necked bottle, then a 10% methanol aqueous solution and 10% Pd/C were added, and hydrogen was introduced. The reactions were performed at 25, 35, and 50°C, respectively. Meanwhile, to investigate the effect of hydrogen pressure, the reactions were performed at 50°C with the hydrogen pressure of 10 psi and 100 psi, respectively. When thin-layer chromatography (TLC) showed that the reaction was complete, the product was washed, dried, and proceeded directly to the next condensation reaction.

Equivalence Ratio of IM-2/IM-3–6/HATU

IM-3–6 was used as a material for the synthesis of dimethyl 3,3′-((((3,6-dioxopiperazine-2,5-diyl)bis(butane-4,1-diyl))bis(azanediyl))bis(carbonyl))bis(bicyclo[1.1.1]pentane-1-carboxylate) (IM-4–6) ([Fig. 11]). The reaction was performed in the amount of IM-2/IM-3–6/HATU equiv. ratio of 1:2:2, 1:3:3, and 1:4:4, and the progress of the reaction was monitored by TLC. After the reaction was completed, a large amount of solid was produced. After filtration, the filter cake was hot-slurried with methanol, and then filtered and dried after slurry cooling to obtain the product IM-4-6.

Fig. 11 Optimization of DBBB6 synthesis process.

Animal Study

To study the potential of DBBB6 in DPI formulations, SMG was used as a model drug, the formulation was optimized, and the SMG-DBBB6 dry powder was prepared through the following procedure. Initially, 200 mg of DBBB6 was uniformly dispersed in deionized water using a 250 mL eggplant-shaped bottle. The suspension was carefully adjusted to pH 7.0 to 7.5 by dropwise addition of 0.1 mol/L aqueous ammonia solution under magnetic stirring. Subsequently, 30 mg of SMG was gradually incorporated into the mixture and allowed to react overnight with continuous agitation. The resulting complex was freeze-dried under vacuum and then pulverized using air flow.

Under Institutional Animal Care and Use Committee approval (IACUC number: SIPI-GL-032), 30 Sprague-Dawley rats (160–170 g, SLAC ANIMAL, Shanghai, China) were randomly assigned to three experimental groups (n = 10 per group) using stratified randomization. The treatment group underwent pulmonary administration of SMG-DBBB6 dry powder aerosols via nose-only inhalation chambers (DBBB6 dose: 1 mg/kg). Control groups received: (1) isodose SMG aerosols formulated with FDKP excipient (vehicle control) and (2) nebulized normal saline (0.9% NaCl, negative control). The number of coughs within 30 minutes after inhalation was recorded using video and sound monitoring (VitaloJAK, Vitalograph Ltd., UK). Each experiment was repeated at least three times. Data are presented as the mean ± standard error of the mean. Student's t-test was used to compare two groups, with p < 0.05 being considered a significant difference. Airway resistance was measured using spirometry (MedGraphics Platinum Elite, MGC Diagnostics, United States).

Conflict of Interest

None declared.

Supporting information

The chemical structure and 3D simulating structures ([Supplementary Table S1], available in the online version), and ¹H NMR, ¹³C NMR, and IR spectra of FDKP and DBBB1-DBBB15 ([Supplementary Figs. S1–S46], available in the online version), can be found in the [Supporting Information] section of this article's webpage.

Ethical Approval

All animal procedures were conducted according to the Chinese legislation and regulations of Laboratory Animals of the Chinese Animal Welfare Committee. The protocols were approved by the Ethics Committee of the Center for Pharmacological Evaluation and Research (Shanghai 200437, China).

• References

• 1 de Boer AH, Hagedoorn P, Hoppentocht M, Buttini F, Grasmeijer F, Frijlink HW. Dry powder inhalation: past, present and future. Expert Opin Drug Deliv 2017; 14 (04) 499-512
• 2 Hebbink GA, Jaspers M, Peters HJW, Dickhoff BHJ. Recent developments in lactose blend formulations for carrier-based dry powder inhalation. Adv Drug Deliv Rev 2022; 189: 114527
• 3 Salústio PJ, Amaral MH, Costa PC. Different carriers for use in dry powder inhalers: characteristics of their particles. J Aerosol Med Pulm Drug Deliv 2024; 37 (06) 307-327
• 4 Abiona O, Wyatt D, Koner J, Mohammed A. The optimisation of carrier selection in dry powder inhaler formulation and the role of surface energetics. Biomedicines 2022; 10 (11) 2707
• 5 Aziz S, Scherlieβ R, Steckel H. Development of high dose oseltamivir phosphate dry powder for inhalation therapy in viral pneumonia. Pharmaceutics 2020; 12 (12) 1154
• 6 Kohlhäufl M, Haidl P, Voshaar T, Häussinger K, Köhler D. [Powder inhalation systems]. Dtsch Med Wochenschr 2004; 129 (39) 2048-2052
• 7 Steckel H, Bolzen N. Alternative sugars as potential carriers for dry powder inhalations. Int J Pharm 2004; 270 (1-2): 297-306
• 8 Celi SS, Fernández-García R, Afonso-Urich AI, Ballesteros MP, Healy AM, Serrano DR. Co-delivery of a high dose of amphotericin B and itraconazole by means of a dry powder inhaler formulation for the treatment of severe fungal pulmonary infections. Pharmaceutics 2023; 15 (11) 2601
• 9 Kadota K. Design of spray-dried porous particles for sugar-based dry powder inhaler formulation. Yakugaku Zasshi 2018; 138 (09) 1163-1167
• 10 Baloira A, Abad A, Fuster A. et al. Lung deposition and inspiratory flow rate in patients with chronic obstructive pulmonary disease using different inhalation devices: a systematic literature review and expert opinion. Int J Chron Obstruct Pulmon Dis 2021; 16: 1021-1033
• 11 Phanstiel IVO, Lachicotte RJ, Torres D. et al. Nanoconstruction of microspheres and microcapsules using proton-induced phase transitions: molecular self-recognition by diamide diacids in water. Chem Mater 2001; 13: 264-272
• 12 Liu N, Li W, Chen H, Wang H. Synthesis and characterization of N-fumaroylated diketopiperazine for pulmonary drug delivery. Mater Express 2023; 13: 39-53
• 13 Rosenstock J, Franco D, Korpachev V. et al; Affinity 2 Study Group. Inhaled technosphere insulin versus inhaled technosphere placebo in insulin-naïve subjects with type 2 diabetes inadequately controlled on oral antidiabetes agents. Diabetes Care 2015; 38 (12) 2274-2281
• 14 Setji TL, Hong BD, Feinglos MN. Technosphere insulin: inhaled prandial insulin. Expert Opin Biol Ther 2016; 16 (01) 111-117
• 15 Gu Q, Wiggers ME, Gleich GJ, Lee LY. Sensitization of isolated rat vagal pulmonary sensory neurons by eosinophil-derived cationic proteins. Am J Physiol Lung Cell Mol Physiol 2008; 294 (03) L544-L552
• 16 Grant ML, Krueger JA, Fineman M. et al. 1350-P: safety and pharmacokinetics of technosphere insulin in pediatric patients. Diabetes 2019; 68: 1350-P
• 17 Mikhail N. Safety of technosphere inhaled insulin. Curr Drug Saf 2017; 12 (01) 27-31
• 18 Available at: https://pubchem.ncbi.nlm.nih.gov/compound/p-nitrobenzaldehyde#section=GHS-Classification
• 19 Oleck J, Kassam S, Goldman JD. Commentary: why was inhaled insulin a failure in the market?. Diabetes Spectr 2016; 29 (03) 180-184
• 20 Lambeir AM, Proost P, Durinx C. et al. Kinetic investigation of chemokine truncation by CD26/dipeptidyl peptidase IV reveals a striking selectivity within the chemokine family. J Biol Chem 2001; 276 (32) 29839-29845
• 21 Watts KS, Dalal P, Murphy RB, Sherman W, Friesner RA, Shelley JC. ConfGen: a conformational search method for efficient generation of bioactive conformers. J Chem Inf Model 2010; 50 (04) 534-546

Publication History

Received: 12 March 2025

Accepted: 25 September 2025

Article published online:
06 November 2025