Doped-MXene assists in deciphering metabolic signature of psoriasis and unraveling dysregulated leukotriene metabolism

2.1 Synthesis of MXene/MWCNTs

Using Ti powder, Al powder, and C powder as raw materials, with a molar ratio of Ti:Al:C = 3:1.2:2.1, the raw materials are mixed uniformly after ball milling for 10–20 h (Eom et al., 2020; J. Zhang et al., 2019; Zhu et al., 2020). The mixture is then sealed in a boron nitride or graphite crucible and placed in a SiC tube furnace with flowing Ar gas as a protective atmosphere. The temperature is ramped up at a rate of 10 ∼ 40 °C/min to reach 1300 ∼ 1450 °C. Subsequently, the mixture undergoes a solid–liquid phase reaction for 0.5 ∼ 4 h at the corresponding temperature. This method of invention offers high product purity, short reaction time, uniform particle size distribution, convenient operation, and a simple process. The synthesized 3.0 g Ti₃ALC₂, 0.01 g sodium fluoride (NaF) and 0.1 g multi-walled carbon nanotubes (purchased from Chengdu Organic Chemistry Co., Ltd., Chinese Academy of Sciences) are dissolved in 10 mL of 1 M sodium hydroxide solution (Shang et al., 2020; Zeng et al., 2020; Zhang et al., 2020). The mixture is subjected to hydrothermal etching under ultrasonic conditions for 8 h. Afterwards, the solution is centrifuged three times, and the corresponding product is collected. The final synthesized MXene/MWCNTs are stored in a refrigerator at 4 °C.

2.2 Animal model of psoriasis

The mouse model induced by imiquimod (IMQ), is widely applied for research purposes (Pinget et al., 2022; Zhang et al., 2021). It not only effectively replicates the cutaneous manifestations of psoriasis-like lesions but also exhibits intrinsic immune dysregulation and acquired immune responses similar to psoriasis. This model is preferred due to its ease of manipulation, high success rate, and extensive utilization in literature. The specific modelling procedure is as follows: Carefully remove the fur from the central area of the mouse's back, creating an exposed area measuring 2 cm × 3 cm. Apply a mild depilatory cream to remove the surface fuzz. Apply 5 % imiquimod (62.5 mg per application for the 2 cm × 3 cm skin area) continuously to the mouse's back skin for 5 to 10 days. Evaluate the skin condition daily and collect serum samples from successfully modelled mice for subsequent psoriasis metabolomics analysis.

2.3 H&E staining

To begin, place the mouse in an appropriate anaesthetic environment and induce anaesthesia using isoflurane. Once the mouse is completely anaesthetized, secure its back using tape or clamps. Proceed to cleanse the skin surface with a disinfectant, ensuring a sterile environment. Gently remove the hair from the central region of the mouse's back using medical scissors and forceps, revealing the underlying epidermal tissue. If needed, cautiously utilize a surgical scalpel to retrieve thin slices of the epidermal tissue. Following this, process the fixed skin sample into sections via deparaffinization and dehydration. Immerse the sections in a hematoxylin solution, resulting in cell nuclei staining in varying shades of purple and blue. Subsequently, engage in destaining and bluing steps to heighten the clarity of nuclear staining. Proceed by staining the sections with the acidophilic eosin dye, imparting pink hues to the cytoplasm and extracellular structures. Lastly, prepare the tissue sections for microscopic examination by undergoing dehydration, clearing, and slide mounting procedures (Lai et al., 2018; Y. Zhang et al., 2019).

2.4 Metabonomics analysis

Metabolomics analysis encompasses techniques such as Principal Component Analysis (PCA), Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA), heatmaps, volcano plots, and enriched pathway identification, facilitated through MetaboAnalyst 5.0 (https://www.metaboanalyst.ca/). The processing of high-throughput mass spectrometry data necessitates preprocessing, peak alignment, and calibration procedures.

2.5 Validation of 5-LO and LCT4s

In the pursuit of mRNA testing for 5-LO and LTC4s, an initial step involved the extraction of total RNA from the samples. This was succeeded by the process of reverse transcription, facilitated by the utilization of a reverse transcriptase enzyme to effectuate the synthesis of cDNA. Following this, primers that held a complementary affinity for the target mRNA sequences were meticulously designed, thereby underpinning the subsequent stages. The preparatory phase included the establishment of specialized PCR reaction systems. Employing a thermal cycler as the vehicle, the amplification phase ensued, characterized by a tripartite temperature progression encompassing denaturation, annealing, and extension. After the amplification process, the resultant PCR products were subjected to meticulous analysis via agarose gel electrophoresis. This analytical method facilitated a discerning comparison with discerning molecular weight markers, the purpose of which was to definitively corroborate the presence of the desired PCR products specific to the designated targets.

2.6 LDI detection and characterizations and XRD analysis

The LDI testing for MXene, MWCNTs, and MXene/MWCNTs involved the utilization of 100 μg/mL concentrations of creatinine and phenylalanine. For assessing protein tolerance, a mixture of standard solutions and 10 mg/mL bovine serum albumin was employed. To evaluate salt tolerance, standard solutions were blended with 5.0 mg/mL NaCl. The XRD analysis was conducted using a Rigaku Miniflex instrument with CuKα radiation (λ = 1.54060 Å) and a scanning rate of 2θ/m. The 2θ range spanned from 5° to 60° during the analysis.

3.1 Morphology and elemental characterization of MXene/MWCNTs

The judicious selection of an appropriate matrix material assumes paramount significance within metabolite detection, particularly in the intricate context of biological samples like serum. In the pursuit of achieving high-throughput capabilities and exceptional sensitivity, the efficacy of matrix materials in ionization and laser transmission assumes a pivotal role in governing the expansive scope of metabolite detection. Therefore, we have chosen MXene materials with a layered structure as the foundational material for LDI metabolite detection. MXene (Fig.S1a-e, Fig.S2, Table S1) exhibited a unique two-dimensional morphology, and its layered structure facilitates the laser desorption ionization of various metabolites on its surface. We also conducted X-ray diffraction (XRD, Fig.S1f) analysis on the synthesized MXene material and found that the peaks at 18.57° and 29.2° correspond to the (0 0 6) and (0 0 8) characteristic peaks, consistent with the literature (Malik et al., 2022), which effectively validated the successful synthesis of the MXene. To further optimize its performance, we selected multi-walled carbon nanotubes (MWCNTs, Fig. 2a) with excellent laser transmission efficiency for modification through a hydrothermal reaction. MXene/MWCNTs displayed a distinctive morphology where multi-walled carbon nanotubes were encapsulated within its layered structure (Fig. 2b). TEM analysis of MXene/MWCNTs (Fig. 2c) revealed a lattice spacing of 0.026 nm, which is consistent with prior literature (Abbasi et al., 2021). Furthermore, we conducted a mapping analysis of MXene/MWCNTs (Fig. 2d, Fig S3), and elemental analysis including C, Al, Ti, and O confirmed the elemental composition (Fig. 2e, Fig. S4). In summary, our SEM and TEM characterizations effectively validated the morphology and elements of MXene/MWCNTs, laying the foundation for future applications.

Close

Fig. 2

Morphology and elemental characterizations of MXene/MWCNTs. a) SEM image of MWCNTs. b) SEM analysis of MXene/MWCNTs displayed a distinctive morphology where multi-walled carbon nanotubes were encapsulated within its layered structure. c) TEM analysis of MXene/MWCNTs. d-e) Mapping analysis of MXene/MWCNTs, including elements C, Al, Ti, and O.

Export to PPT

3.2 Performance evaluation of MXene/MWCNTs for LDI-MS

The meticulous assessment of the matrix material's performance in metabolomic analysis plays a pivotal role, particularly when delving into the intricacies of psoriasis. Consequently, a thorough evaluation was undertaken to comprehend the matrix's efficacy. As illustrated in Fig. 3a-b, the successful identification of phenylalanine and creatinine using MXene/MWCNTs, accompanied by the emergence of discernible sodium and potassium peaks, unequivocally affirmed the potential of MXene/MWCNTs in detecting standard samples. With meticulous refinement, integrating MWCNTs into MXene demonstrated elevated sensitivity. Specifically, the peak intensity of MXene/MWCNTs (2635 ± 318, Fig. 3c) increased by 60.6 % compared to pure MXene (3547 ± 498) and 64.3 % compared to pristine MWCNTs (2587 ± 357). Moving onward, the capacity of MXene/MWCNTs to discern an increased number of peaks within clinical sample detection was unveiled, an invaluable attribute when confronted with the intricacy of analyzing multifaceted clinical samples. Especially, the number of peaks for MXene/MWCNTs increased from 70 ± 10, as depicted in MXene, and 62 ± 9, as seen in MWCNTs, to 88 ± 11 (Fig. 3d). Furthermore, MXene/MWCNTs exhibited exceptional tolerance to proteins (Fig. 3e) and salts (Fig. 3f), fortifying its utility. The comprehensive validation of standard samples, optimization, protein tolerance, and salt tolerance collectively underscored the practicality of MXene/MWCNTs as an LDI matrix material, subsequently laying the groundwork for upcoming psoriasis metabolomic analyses.

Close

Fig. 3

Performance evaluation of MXene/MWCNTs for LDI-MS. Performance of MXene/MWCNTs as LDI matrix for detecting a) phenylalanine and b) creatinine. c) Detection intensity statistics for MXene, MWCNTs, and MXene/MWCNTs, and d) Peak number statistics for MXene, MWCNTs, and MXene/MWCNTs in clinical samples. The detection performance of MXene, MWCNTs, and MXene/MWCNTs for standard samples includes e) protein tolerance and f) salt tolerance.

Export to PPT

3.3 LDI-MS-based metabolomic fingerprinting of psoriasis using MXene/MWCNTs

In the pursuit of comprehending the intricate tapestry of psoriasis, it becomes imperative to discern it from the backdrop of control groups. This elucidation provides a gateway to uncover the unique metabolic perturbations intrinsic to psoriasis. Accordingly, we embarked on a multifaceted exploration, encompassing skin and animal model serum biosamples, thereby approaching psoriasis from diverse angles. In Fig. 4a, we delineated our concerted efforts to gather skin samples from healthy individuals and psoriasis patients (Table S3). Blood extraction through meticulous scraping sets the stage for the subsequent metabolomic analysis. Employing a judicious combination of principal component analysis (PCA) and orthogonal partial least squares discriminant (OPLS-DA), as showcased in Fig. 4a-b, we efficaciously discerned between the skin samples sourced from 32 healthy individuals and those from 38 individuals grappling with psoriasis. These distinctions paved the way for the embodiment of molecular alterations in the form of upregulated and downregulated shifts, expertly presented through volcano plots in Fig. 4c. These plots serve as intuitive visualizations, encapsulating the essence of metabolite variations under distinct conditions. Akin to a symphony of data, heatmaps in Fig. 4d unfold a mesmerizing visual panorama that accentuates the intricate metabolic disparities between skin samples originating from healthy individuals and psoriasis patients. These heatmaps, adeptly deciphering complex multidimensional information, offer a window of insight into potential biomarkers and biological interconnections, augmenting our comprehension of the meaningful significance of metabolites. Through these endeavours, the synergy of methodologies such as OPLS-DA, volcano plots, and heatmaps contributes profoundly to the intricate analysis and discovery of biomarkers within psoriasis skin samples, thereby unravelling pivotal insights into the molecular underpinnings governing the evolution and progression of this enigmatic malady.

Close

Fig. 4

LDI-MS-based metabolomic fingerprinting of clinical psoriasis skin samples using MXene/MWCNTs. a) PCA and b) OPLS-DA analyses distinguishing control and psoriasis groups. c) Volcano plot analysis of 32 control and 38 psoriasis skin biosamples, with blue and red representing downregulated and upregulated molecules, respectively. d) Heatmap analysis of control and psoriasis groups, highlighting the differences between the two groups.

Export to PPT

Transitioning our exploration, we delve into the realm of serum biosamples from the psoriasis animal model, harnessing Laser Desorption/Ionization Mass Spectrometry (LDI-MS) to unearth metabolic fingerprints. The establishment of the psoriasis animal model, carefully executed through imiquimod application, is graphically represented in Fig. 5a. The transformative impact of this approach on the epidermal landscape is vividly portrayed in histopathological images presented in Fig. 5b, corroborating the successful establishment of the psoriasis model. Intricacies of infiltrative erythema, silvery-white scales, and scattered bleeding spots encapsulate the visual cues of psoriasis-like manifestations on the murine canvas post-model creation. In a symphony of analytical techniques, Fig. 5c-d resonated with the successful differentiation of serum samples originating from 32 control mice and 32 mice representing the psoriasis model. The ensuing crescendo of potential biomarkers found visual embodiment through volcano plots (Fig. 5e) and heatmaps (Fig. 5f), echoing our resolute commitment to dissecting psoriasis-specific metabolic variances at the animal model stratum.

Close

Fig. 5

MXene/MWCNTs LDI-MS-based metabolomic fingerprinting of serum biosamples of psoriasis animal model. a) Establishment of the psoriasis animal model through imiquimod application. b) Histopathological images of mouse skin before and after model establishment, highlighting the impact of imiquimod on epidermal texture and confirming successful model creation. c) PCA and d) OPLS-DA analyses distinguishing control and psoriasis groups among animal model samples. e) Volcano plot analysis of 32 control and 32 psoriasis animal model serum biosamples, with blue and red representing downregulated and upregulated molecules, respectively. f) Heatmap analysis of control and psoriasis groups, highlighting the differences between the two groups.

Export to PPT

3.4 Diagnostic performance and biomarker analysis of psoriasis

Fig. 6a-B reflected our meticulous diagnostic performance evaluation viaReceiver Operating Characteristic (ROC) curve analysis, a strategy that scrutinizes the sensitivity and specificity of classification models across varied thresholds. Impressively, our metabolomic approach presented robust diagnostic efficacy in clinical and animal model samples, epitomized by Area Under the Curve (AUC) values of 0.959 and 0.924, respectively. This resounding endorsement attests to the potency of our methodology in bestowing diagnostic precision to psoriasis. Venturing further into the realm of metabolite analysis, we spotlight two key players-arachidonic acid and leukotriene E4. Fig. 6c and Fig. 6e showcased the elevation of these metabolites in psoriasis patients, their diagnostic potential further bolstered by impressive AUC analyses (Fig. 6d and Fig. 6f).

Close

Fig. 6

Diagnostic performance and biomarker analysis of psoriasis using LDI-MS-based metabolomic approach with MXene/MWCNTs. a) Diagnostic performance of clinical samples, and b) diagnostic performance of animal model samples. Analysis of c) arachidonic acid intensity and d) area under the curve (AUC) in control and psoriasis groups. Analysis of e) leukotriene E4 intensity and f) AUC in control and psoriasis groups.

Export to PPT

3.5 Validation of biomarkers and analysis of metabolic pathways

In pursuit of more profound insights, we gaze upon the intriguing landscape of metabolic pathways, particularly the role of arachidonic acid and its intricate cascade. As revealed through PCR amplification, the expression of 5-LO and LTC4s in the psoriasis group underscores the dynamic role of the arachidonic acid pathway. This pathway serves as a conduit for transforming arachidonic acid into bioactive entities, orchestrating inflammation and immune modulation. In a harmonious choreography, LTC4s emerge as a pivotal player, converting arachidonic acid into leukotriene E4, a potent member of the leukotriene family implicated in processes ranging from inflammation to cellular signalling. Notably, the heightened expression of 5-LO and LTC4s within the psoriasis group amplifies the salience of the arachidonic acid pathway in psoriasis (Fig. 7a). In the panoramic vista of enrichment pathway analysis, as depicted in Fig. 7b, the arachidonic acid pathway emerges as a prominent presence, intricately interwoven with a constellation of metabolic pathways. This complex interplay elucidates the pivotal role of arachidonic acid and its associated pathways in the psoriasis narrative (Fig. 7c), affording profound insights into its occurrence, progression, and potential avenues for therapeutic intervention.

Close

Fig. 7

Validation of biomarkers and exploration of related metabolic pathways. a) mRNA expression levels of 5-LO and LTC4s in control and psoriasis groups. b) Enrichment pathway analysis associated with psoriasis, and c) corresponding metabolic pathway network, with arachidonic acid significantly highlighted and exhibiting significant p-values.

Export to PPT