RP-HPLC-DAD analysis

RP-HPLC analysis of DAL and the exposed samples was performed. The DAL isoforms A0, A1, B0, B1 and B2 eluted at 15.66 min, 16.48 min, 20.28 min, 21.03 min and 21.56 min with a resolution of 1.86, 8.75, 1.93 and 1.96, respectively. System suitability was performed with the DAL standard working solution (200 ppm) and checked before each analysis to address the parameters concerning the chromatographic acceptance criteria. System suitability criteria such as retention time, theoretical plates (N), resolution and peak asymmetry factor were found to be more than 2000, greater than 1.8, and less than 1.2, respectively (S−1).

Photolytic studies

The photolytic effects on DAL were studied by exposing the sample to UV and visible light. The chromatographic analysis of the samples exposed to UV light (3 cycles, 200 W.h.m−2) did not show any degradation products however, the peak response of DAL was reduced after the exposure to each UV cycle. The decrease in response of DAL is due to the formation of either non-polar degradation products, which are insoluble in water, as the diluent system is water or the adsorption of DAL to the sample container. Two approaches were incorporated to address this finding: (i) incorporation of different diluent systems such as ACN (aprotic), DMSO (neutral, aprotic), and MeOH (protic); (ii) to check the adsorption of DAL on the surface containers followed by UV exposure, as peptides have the tendency to adsorb on the surfaces of containers [27]. The HPLC analysis for the samples exposed to visible light (1.2 million lux.h) was performed and HPLC analysis revealed four degradation products (DP)—VDP−1, 2, 3, 4 (Table 1)—with their RTs at 4.88, 8.91, 16.47 and 19.34 min, respectively (S-2).

Table 1 Exposure conditions of DAL along with their respective HPLC chromatographic detailsApproach I: Effect of diluent system on DAL upon exposure to UV

Samples were collected after each UV cycle with ACN, MeOH and DMSO as diluent systems apart from water and proceeded with HPLC analysis. No other peaks apart from DAL and its isoforms were eluted with ACN, MeOH and DMSO as diluent systems after reconstitution. This result strongly suggests no other insoluble degradation product was formed upon exposure to UV light and DAL is relatively stable upon exposure to UV conditions. However, DAL was adsorbed on the surfaces of the containers when exposed to UV, which was confirmed through the HPLC analysis after reconstitution with diluents of diverse nature.

Approach II: Adsorption studies via FE-SEM and EDAX analysis

The DAL samples followed by the UV exposure and the control samples were examined through FE-SEM. Characteristic results in Fig. 1 show the FE-SEM micrographs of UV light control (three details; (a, b, and c)) and the DAL UV-exposed sample (three details; (d, e, and f)) (c and f are respective images under higher magnification.). These investigations revealed clear surfaces in the control samples where the presence of no particulate content on the surface of the container was seen. However, in the sample exposed to UV, the presence of particles on the surfaces of the glass was seen, distinguishing the phenomenon of adsorption. The morphology of the adsorbed particle was identified to be a globular structure (Fig. 1f). The formation of the globular structure was due to the exposure of hydrophobic pockets of DAL when treated under UV, leading to the formation of micelles, a self-assemble phenomenon. When DAL was UV irradiated, the amino acid residues would have been activated through energy absorption and caused modifications in their inherent conformation, leading to either unfolding or rearrangements. The unfolding of amino acids has exposed the hydrophobic regions, which were buried inside in DAL, towards the outer surfaces. The electrostatically active outer hydrophobic regions interact with each other, thus initiating the formation of self-assembly [28]. Based on the inherent properties, the self-assembled peptide may obtain various shapes where the self-assembled DAL was deciphered to be a globular-like structure (particle size ranging from 100 to 500 nm), which was confirmed through FE-SEM investigations.

Fig. 1

Details of SEM images of DAL UV control and UV exposed glass surface. a UV glass control. b Surface of UV glass control under 5 K magnification. c Surface of UV glass control under higher magnification (25 K). d DAL UV exposed glass. e Surface of DAL UV exposed glass under 5 K magnification. f Surface of DAL UV exposed glass under higher magnification (25 K)

Through energy-dispersive X-ray analysis (EDAX), mapping and quantitative studies were performed to investigate the distribution and relative quantification of DAL adsorbed on the glass surfaces. Through these studies, the percentages of carbon (C), nitrogen (N), oxygen (O) and chlorine (Cl) were majorly obtained, which are the core constituents of DAL. Apart from these, other elements such as silica (Si), aluminum (Al), sodium (Na), potassium (K), and calcium (Ca) were also seen during EDAX studies corresponding to the native composition of the glass (Fig. 2). In the UV light control samples, the C (w %), N (w %), and O (w %) were found to be 0.63, 1.05, and 47.17, respectively. However, in the sample exposed to UV light, the C (w %), N (w %), O (w %), and Cl (w %) were 27.03, 3.19, 31.67, and 0.84, respectively. The increase in C (w%), N (w%), O (w %), and Cl (%) in the UV-treated sample against the control is attributed to the deposition of DAL on the surface of the glass container. Furthermore, the presence of Cl in EDAX analysis has confirmed the phenomenon of adsorption, as Cl is the core unique constituent present only in the chemical composition of DAL. The relative quantitation details of the elements obtained through EDAX analysis were provided in S-3. Furthermore, the incurred phenomenon is evident through the HPLC quantitative results, where the DAL assay in the UV-exposed sample was reduced to a level of 43%.

Fig. 2

EDAX analysis for a DAL UV control sample and b DAL UV exposed sample

MS and tandem mass studies of visible DPsMS/MS fragmentation behavior of VDP-1

Degradation of DAL upon exposure to visible light led to the formation of VDP-I with a retention time of 4.54 min. After the LC-MS analysis, the MS spectra exhibited a molecular ion peak with m/z 899.3284 (2), with a mass difference of 17.9663 Da from DAL. The deconvoluted mass for VDP-1 was 1796.6414 with a molecular formula of C88H101ClN10O29 (Fig. 3). The most likely explanation for the possible mass difference is the removal of the Cl group and the addition of the OH group to the aa 3 in the core peptide. The MS/MS spectra exhibited the major ions with m/z 890.3231 (2), 874.8022 (2) and 712.7238 (2), with its singly charged m/z 1424.4384 (1). Apart from these masses, m/z 1393.3966 (1), m/z 698.7264 Da (2) and m/z 616. 1762 (2) were also present. The m/z 890.3231 (2) was formed from VDP-I by the loss of the water molecule from the aa 6 of the core peptide. This fragment, with the further loss of the CH5N group from the N-terminus, has given a fragment with an m/z 874.8022 Da (2). The loss of the glycolipid moiety from the aa 4 of the parent fragment 899.3284 (2) generated m/z 1425.4397 (1) and m/z 712.7238 (2), which is evident by the presence of mass in the MS/MS spectra. A simultaneous loss of CH5N from the N-terminus and CO from the G2 of m/z 1425.4397 (1) has given m/z 1393.3966 (1) and 698.7264 (2), respectively. The mass with m/z 1393.3966 (1) further fragmented to give m/z 616. 1762 Da (2) by the glycolytic cleavage of M2 from m/z 1393.3966 (1). The identified masses, along with their ppm error, are provided in Table 2. The comprehensive fragmentation of VDP-1 was detailed in S-4.

Fig. 3

X, photolytic; Y, thermal; 1, total ion chromatogram (TIC); and 2, extracted ion chromatogram (EIC) of various masses for DAL under respective conditions

Table 2 Identified HRMS and MS/MS fragments with their mass ppm error for DAL and DPsMS/MS fragmentation behavior of VDP-2

The degradation product VDP-2 eluted at 7.54 min. The MS analysis revealed the molecular ion peak with m/z 890.3234 (2), with a deconvoluted mass of 1778.6312 and molecular formula of C88H99ClN10O28. The mass difference of VDP-2 from DAL was identified as 33.9604 Da. The probable way for the addition of 33.9604 Da to DAL is due to the formation of an epoxide on the aa 4. To comprehend the proposed mass loss further, MS/MS analysis of m/z 890.3234 (2) was performed, where major daughter ions with m/z 1393.3944 (1), m/z 874.8016 (2), m/z 834.7962 (2), and m/z 787.2769 (2) were identified. The masses with m/z 876.3253 (2), m/z 865.7963 (2), m/z 856.7897 (2), m/z 843.8013 (2), m/z 793.7794 (2), m/z 784.7691 (2), and m/z 778.2722 (2) were also present in the MS/MS spectra. The loss of a CO group from the G2 led to the formation of m/z 876.3253 (2), and the simultaneous loss of CH5N from the VDP-2 has given an m/z 874.8016 (2). These two daughter fragments serve as the characteristic fragments confirming their losses from the parent fragment of m/z 890.3234, which is only possible when the mass difference is 33.9604 Da, apprehending the loss of Cl. The mass of m/z 874.8016 (2) was further fragmented to m/z 1393.3944 (1) and 865.7963 (2) by the loss of glycolipid and water respectively. The simultaneous losses from m/z 865.7963 (2) generated fragments with m/z 856.7897 (2) by the loss of a water molecule from the M1 on the aa 4, m/z 843.8013 (2) with the loss of CO from M2 present on aa 7 and 793.7742 (2) by the complete loss of M2 from aa 7. The fragment with m/z 843.8013 (2) further underwent water loss from M1, giving m/z 834.7962 (2). The simultaneous loss of a water molecule and NH from m/z 793.7742 (2) incurred, giving m/z 787.2769 (2) and m/z 784.7691 (2), respectively. The fragment with m/z 787.2769 (2) further lost a water molecule, which is confirmed by the presence of m/z 778.2722 (2) in MS/MS spectra. The detailed plausible fragmentation pathway along with their respective mass losses was comprehended in S-5. The identified masses along with their ppm error are provided in Table 2.

MS/MS fragmentation behavior of VDP-3

VDP-3 has a retention time of 13.34 min. The MS spectra have shown a molecular ion peak similar to VDP-2 with an m/z 890.3234 (2) with a deconvoluted mass of 1778.6310 and a molecular formula of C88H99ClN10O28. A similar MS/MS fragmentation pattern of VDP-3 as of VDP-2 was observed during the MS/MS analysis. The identical fragmentation pattern in MS/MS studies and the elution pattern obtained on RP-HPLC C18 column suggested that VDP-3 (Iso VDP-2) is a diastereomeric product of VDP-2 formed within the structure. As the cyclic peptide core of VDP-2, 3 would be the same, a change in the residue 3 Cα configuration would cause the peptide ring to undergo comparable conformational changes, exposing identical molecular regions to the stationary phase and solvent. As both the diastereomeric products have different properties and interact with the stationary phase in a different manner, the elution of diastereomers had occurred lately. The daughter ions with m/z 834.7990 (2), m/z 874.8018 (2), and m/z 876.3268 (2) were the common fragments identified in both VDP-2 Iso VDP-2. The MS/MS analysis generated fragments with m/z 874.8018 (2) by the loss of CH5N from the N-terminus. The loss of C2H5NO2 and water from the glycolipid part of Iso VDP-2 led to the formation of m/z 834.799 (2). The m/z 876.3268 (2) was formed by the loss of CO from Iso VDP-2, which was further fragmented, giving an m/z 698.7254 Da (2) by the loss of glycolipid on aa 4. The glycolipid part was lost from Iso VDP-2 directly, which is evident by the presence of masses with m/z 712.7233 (2) and m/z 1424.4382 (1). This loss was followed by the elimination of CH5N leading to the generation of m/z 1393.3965 (1). The loss of M2 from m/z 698.7254 (2) through hydrolysis of the O-glycosidic linkage occurred, which further produced a fragment with m/z 616.1760 (2). The comprehended fragmentation pattern and identical mass losses are provided in S-6. The identified masses along with their mass ppm error are provided in Table 2.

MS/MS fragmentation behavior of VDP-4

The degradation product VDP-4 eluted at 17.96 min. LC-MS analysis was performed revealing the mass of m/z 891.3306 (2) in the MS spectra. The deconvoluted mass was tailored to be 1780.6456 with a molecular formula of C88H101ClN10O28. A mass difference of 33.9612 Da was identified due to the loss of the Cl moiety from the peptide core. In MS/MS analysis of m/z 891.3306 (2), ions with m/z 877.3316 (2), 712.7232 (2) with its singly charged m/z 1424.4391 (1), m/z 1393.3965 (1), and m/z 616.1759 (2) were majorly identified. Also, masses with m/z 1231.3444 (1), m/z 698.7262 (2), and m/z 602.1770 (2) were also seen in the MS/MS spectra. The fragment of m/z 712.7232 (2) was formed due to the loss of the glycolipid moiety from the aa 4. This mass was further fragmented to m/z 1393.3965 (1) by the loss of CH5N from the N-terminus. The loss of CO from M2 (hemi acetal) has given a fragment with m/z 877.3316 (2). The subsequent loss of glycolipid from m/z 877.3316 (2) was evident by the presence of mass m/z 698.7262 (2). The m/z 616.1759 (2) with its singly charged m/z 1231.3444 (1) was formed either by the removal of M2 from aa 7 of m/z 1393.3965 (1) or by the loss of CH5N and C5H12O4 from m/z 698.7262 (2). Also, the presence of m/z 602.1770 (2) evidently demonstrates the loss of water from the precursor fragment of m/z 616.1759 (2). The detailed molecular fragmentation pathway comprehending their respective mass losses is detailed in S-7. The comprehended mechanistic pathway for the formation of VDP-1, 2, 3 and 4 is provided in S-8. The identified masses and their ppm error are provided in Table 2.

Thermal studies

A thermal degradation study was performed at 60 °C in a hot air oven for 48 h. Aliquots of samples were collected at multiple time points and were analyzed using HPLC. The optimum degradation was observed after 28 h (Table 1). The chromatographic data obtained from RP-HPLC indicate the formation of four DPs (TDP-1, 2, 3, 4) with the RT at 17.42, 19.16, 23.58 and 24.34 min, respectively (S-9).

MS and MSn studies

Through MS studies, the masses of the DAL and TDPs were identified as m/z 908.3205 (2) with a deconvoluted mass of 1814.6176 and 1814.6176 ± 0.0012 with a molecular formula of C88H100Cl2N10O28, respectively. The identified mass was similar to that of DAL B0 and B1, which is evidence for the formation of isomeric degradation products (Fig. 4). MS/MS studies confirmed the further fragmentation of m/z 908.3205 (2). The fragmentation of the DPs was similar to the fragmentation pattern of DAL, suggesting the formation of isomeric DPs. However, the intensities and the majorly formed daughter ions are different for the individual isomeric DPs concerning DAL B0 and B1. The intensities of the m/z 1458.4060 (1) and m/z 1427.4060(1) have shown variations in the MS/MS spectra of DPs and DAL B0 and B1. Plausible neutral losses and the varying correlation of the intensities from MS/MS analysis directed to the most plausible sites of isomerization leading to the formation of diastereomers. Further MSn studies for DAL and DPs are elucidated in S-9.

Fig. 4

MS2 of various isomers in DAL thermal exposure

R-value

Tandem mass spectrometry (MS/MS) was employed to investigate the fragmentation route of DAL, its isomeric forms, and the corresponding TDPs. DAL and the TDPs have shown a similar fragmentation pathway; however, prominent differences were perceived in the relative intensities of key daughter fragments of m/z 1458.4063 (a1), 729.7079 (a2) and 1427.3629 (b). Hence, to quantify these differences, an R-value was calculated, defined as the ratio of the intensity of diagnostic fragment b to that of fragments a1 and a2 (Table 3). The calculated R-values for TDP-1, TDP-2, TDP-3 and TDP-4 were 0.74, 0.18, 0.08 and 0.12, respectively, indicating variations in the dominance of their specific fragmentation pathways. A higher R-value was observed for TDP-1 and TDP-2, which suggests that fragments b are more prominent, implying a structural modification that promotes their formation. In contrast, a lower R-value was identified for TDP-3 and TDP-4, indicating a greater abundance of fragment a1/a2, plausibly suggesting an alternative isomeric site of modification. These results collectively suggest that TDP-1 and TDP-2 likely possess distinct isomeric configurations, as reflected by their markedly different R-values, whereas TDP-3 and TDP-4 may share a similar modification site, which is apparent through a minimal difference in their comparable R-values.

Table 3 Calculated R-values for distinguishing daughter fragments for DAL and isomeric DPs with respective their obtained intensities in MS/MS spectraEnergy minimization approach

In correlation with the data obtained from tandem MS studies and R-values, a total of 70 energy minimizations were conducted for different plausible structures with the modifications involving bond stretching, angle bending, torsional energy and non-bound weak interactions for the atomic coordinates [29] for DAL B0, B1 and the TDPs to spot the plausible isomer (S-10). The bond angles and the atomic coordinates (R/S, S/R) were modified at the N-terminus and the respective carbon (MC1, MC2, MC3, MC4, MC5, MC6) present in the glycan (mannose (M)) of DAL to attain the most stable form. Each individual bond at the respective carbon in M was altered by setting the rest of the bonds constant. These altered structures were tailored to identify the most plausible isomeric sites forming TDP-1, 2, 3 and 4 through the energies obtained (Table 4). Through these studies, MC1(S) and MC2(R) were identified as the most plausible sites of isomerization for TDP-1 and 2, producing a stable configuration with energies of −1.31 and 0.05, respectively. Apart from these, the modified N-termini of DAL B0 and B1 were postulated as the rest of the plausible isomeric sites producing TDP-3 and 4, which attained a stable configuration through the flipping of lone pairs of electrons on nitrogen (Fig. 5). The obtained energies for all plausible modifications are provided in S-6. The inversion of lone pairs of electrons on nitrogen in the isomeric N-terminus occurs at higher energies only, which is driven through heat. Moreover, the elution pattern of TDPs on the RP-C18 column also suggests the proposed modifications. The TDPs-1 and 2 have been eluted prior to the B0 peak, comprehending the modification in glycan. As the modified glycan is comparatively more polar due to the change in configuration of the glycan (R/S to S/R), the isomeric TDPs-1 and 2 have routed to a different interaction with the column, thus making it elute early. Similarly, the modified N-terminus in TDPs-3 and 4 has altered the nature of interaction on the RP-C18, making it comparatively less polar, leading to a later elution than B0 (Fig. 6).

Table 4 Energies obtained for various structures through minimization approachesFig. 5

Mechanism of formation of DAL isomeric TDP-1 and TDP-2 upon thermal exposure

Fig. 6

Mechanism of formation of DAL isomeric TDP-3 and TDP-4 upon thermal exposure