Surface-Enhanced Raman Spectroscopy in Urinalysis of Hypertension Patients with Kidney Disease

doi:10.21203/rs.3.rs-3304678/v1

Download PDF

Article

Surface-Enhanced Raman Spectroscopy in Urinalysis of Hypertension Patients with Kidney Disease

https://doi.org/10.21203/rs.3.rs-3304678/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 06 Feb, 2024

Read the published version in Scientific Reports →

You are reading this latest preprint version

Arterial hypertension (AH) is a multifactorial and asymptomatic disease that affects vital organs such as the kidneys and heart. Considering its prevalence and the associated severe health repercussions, hypertension has become a disease of great relevance for public health across the globe. Conventionally, the classification of an individual as hypertensive or nonhypertensive is conducted through ambulatory blood pressure monitoring over a 24-h period. Although this method provides a reliable diagnosis, it has notable limitations, such as additional costs, intolerance experienced by some patients, and interferences derived from physical activities. Moreover, some patients with significant renal impairment may not present proteinuria. Accordingly, alternative methodologies are applied for the classification of individuals as hypertensive or nonhypertensive, such as the detection of metabolites in urine samples through liquid chromatography or mass spectrometry. However, the high cost of these techniques limits their applicability for clinical use. Consequently, an alternative methodology was developed for the detection of molecular patterns in urine collected from hypertension patients. In this study, a direct discrimination model for hypertensive and nonhypertensive individuals was generated through the amplification of Raman signals in urine samples based on gold nanoparticles and supported by chemometric techniques such as partial least squares-discriminant analysis (PLS-DA). Specifically, 162 patient urine samples were used to create a PLS-DA model. These samples included 87 urine samples from patients diagnosed with hypertension and 75 samples from nonhypertensive volunteers. The PLS-DA model with 4 latent variables (LV) was used to classify the hypertensive patients with a calibration sensitivity (SenCal) of 89.2%, cross-validation sensitivity (SenCV) of 75.4%, prediction sensitivity (SenPred) of 86.4%, calibration specificity (SpeCal) of 86.0%, cross-validation specificity (SpeCV) of 77.2%, prediction specificity (SpePred) of 77.8%, calibration accuracy (AccCal) of 87.7%, cross-validation accuracy (AccCV) of 77.0%, and prediction accuracy (AccPred) of 82.5%. This study demonstrates the ability of surface-enhanced Raman spectroscopy to differentiate between hypertensive and nonhypertensive patients through urine samples, representing a significant advance in the detection and management of AH.

Biological sciences/Molecular biology

Health sciences/Biomarkers

Health sciences/Health care

Health sciences/Molecular medicine

Health sciences/Nephrology

Physical sciences/Nanoscience and technology

Hypertension

Urine

Surface-Enhanced Raman Spectroscopy

Nanoparticles

Kidney damage

Chemometrics

Arterial hypertension (AH) is defined as persistently elevated blood pressure, such as systolic blood pressure of ≥ 130 mmHg and/or diastolic blood pressure (DBP) of ≥ 90 mmHg, taken as an average of three correctly measured readings^1–3. Furthermore, AH has been closely associated with different alterations, considering the intrinsic systemic nature of blood circulation, which frequently affects the kidney to lead to the development of kidney diseases (KDs)⁴ that encompass a great heterogeneity of pathophysiological processes that often coexist and overlap with others. Hypertension and KD are interrelated, and both can be the cause and/or consequence of each other⁵.

AH can induce kidney damage through damage to the blood vessels, sodium dysregulation, increased sympathetic nervous system, scar formation, and hardening of the glomeruli. At the cellular level, hypertension can induce significant metabolic changes that contribute to these pathological effects. These changes may involve increased production of reactive oxygen species, the activation of the renin–angiotensin–aldosterone system, and the alteration of the levels of various metabolites. These metabolic changes can trigger the accumulation of toxic waste products and the dysfunction of kidney cells⁶. In several cases, most people with AH do not present signs or symptoms⁷, thereby making it a silent disease. AH is a global public health concern because the number of cases worldwide has increased from 650 million to 1,280 million in the last 30 years^8, and approximately one-third of adults are estimated to have hypertension². Moreover, as per the World Health Organization, an alarming 46% of hypertensive adults remain unaware of their condition⁷, and tragically, 8.5 million deaths have been attributed to the association of hypertension with other comorbidities⁸.

Although the AH diagnosis is simple (it is made through conventional blood pressure measurements in a medical office or through ambulatory blood pressure monitoring), the devices for such measurements must be validated according to standardized conditions and protocols. In addition, the possible masking effect due to the white-coat effect, the discomfort in some patients due to the follow-up time, and the limited availability of access to health systems make it challenging to identify hypertension in certain patients. Therefore, technological tools that can facilitate the classification of hypertensive patients through urinalysis in populations at a higher risk of developing kidney damage due to hypertension can contribute to the clinical characterization of the population to support decision-making. This study developed a methodology for classifying hypertensive patients by analyzing urine samples using surface-enhanced Raman spectroscopy (SERS) coupled with gold nanoparticles (AuNPs) and supported with multivariate statistical analysis (chemometrics).

Raman spectroscopy is an optical spectroscopic technique based on inelastic scattering, which is defined by the shifting energy. This shift is calculated by the difference between the energy of incident photons and the energy of emitted photons, which is equivalent to the vibrational mode energy of the interrogated molecules⁴. Raman spectroscopy offers the advantages of minimal treatment of samples, minimal interference by water molecules, and nondestructive and automatable sample analysis, among others⁹. For this reason, the direct analysis of urine samples is of great interest. Nonetheless, the Raman phenomenon suffers from a disadvantage in that its occurrence probability is weaker than that of other techniques, such as infrared spectroscopy. However, this limitation can be overcome by employing SERS. The implementation of AuNPs is useful owing to their plasmonic properties that enhance the Raman signal on the order of 10⁴–10⁹ times, thereby providing greater sensitivity. In addition, multivariate statistical analysis provides evidence of pattern differences in the spectral signals from complex matrices such as urine samples.

The applications of Raman spectroscopy have attracted interest as a promising alternative technique to address certain disadvantages in diagnosing pathologies⁹. For example, SERS methodologies have been used for the detection of different types of cancer from urine samples, such as breast cancer¹⁰, prostate cancer¹¹, and colorectal cancer^12, and conventional Raman spectroscopy has been used to detect cervical cancer¹³. Other studies have used Raman spectroscopy for the diagnosis of kidney damage by evaluating metabolites from urine samples, such as albumin for diabetic patients¹⁴; creatinine, urea, and glucose for hypertensive diabetic patients¹⁵; hydroxybutyrate, alanine, creatinine, and porphyrins for the diagnosis of renal failure¹⁶; and SERS analysis to determine proteinuria from urine samples¹⁷. In particular, SERS with silver nanoparticles has been used to determine hypertension in blood samples. It evaluates the changes in erythrocytes generated by hypertension¹⁸. The biochemical analysis of urine samples is performed for the classification and diagnosis of healthy people and those with diabetes and hypertension by using Raman spectroscopy¹⁹, and the noninvasive and prospective diagnosis of coronary heart disease from urine samples is performed by using SERS based on the detection of platelet-derived growth factor-BB²⁰.

In this study, we have proposed a direct model for the identification of hypertensive patients through enhanced Raman signals in urine samples coupled with AuNPs, supported by the chemometrics technique Partial Least Squares-Discriminant Analysis (PLS-DA). This alternative methodology for molecular pattern detection in urine samples associated with hypertension allows the distinguishing of hypertensive from nonhypertensive patients based on the spectral patterns obtained from the urine samples. The PLS-DA models were validated and evaluated for their sensitivity, specificity, and accuracy, among other variables.

2.1. Materials

Hydrogen tetrachloroaurate (III) trihydrate (HAuCl₄.3H₂O) ≥ 99.9%, Sigma‒Aldrich (St Louis, MO USA); sodium citrate dihydrate, granular (C₆H₅Na₃O₇.2H₂O) 99.3%, J. T. Baker–Fisher Scientific (Edo. de Mex. Mexico), MiliQ water Arium Comfortm, Sartorius AG (Göttingen, Germany); CLARIO Starplus spectrophotometer (BMG-LABTECH The Microplate Reader Company, Ortenberg Germany); Zetasizer LAB (Malvern Panalytical a spectris company, United Kingdom). Raman 785 L, Wasatch Photonics (Orlando, FL, USA).

2.2. Patients, sample collection, and storage

The Ethics Committee of Universidad Simón Bolívar, Barranquilla–Colombia, approved this study. In addition, written informed consent was obtained for the collection of urine samples from all study subjects (approximately 30 mL for each patient). A total of 162 urine samples were collected from volunteer patients who attended the clinic (Clinica de la Costa, Barranquilla–Colombia), including 87 AH patients and 75 healthy volunteers. The collected urine samples were identified and labeled AH_N° patients for those with a diagnosis of AH and HV_N° patients for healthy volunteers. Then, the urine samples were separated in aliquots of 500 µL and stored at − 80°C until further analysis. In the AH group, 35 patients were diagnosed with kidney damage and were further classified into a subgroup termed RAH. Of the 162 urine samples, 122 were used to generate the prediction model, and 40 were used for external validation. The separation of the samples for the training (to generate the prediction) and external validation of the model was performed randomly, assigning 75% for training and 25% for the validation of the model.

2.3. Synthesis and characterization of AuNPs

AuNPs were synthesized following the protocol described by Hermanson et al. (2013)²¹, albeit with some modifications. AuNPs were prepared via chemical reduction of 20 mM HAuCl₄(III) solution using sodium citrate 2% (w/v). Briefly, 1,250 µL of 20 mM HAuCl₄ was added to 100 mL of boiling Type 1 water under constant stirring at 400 rpm. Then, 500 µL of sodium citrate 2% (w/v) was added to the solution and stirred continuously for 30 min. The color change of the solution indicated the formation of monodisperse colloidal gold particles. The AuNPs were further characterized by determining their absorption maximum with a UV–Vis scan at 350–800 nm (CLARIO Starplus) and by determining the hydrodynamic radius of the colloidal solution through dynamic light scattering (Zetasizer)^22–25.

2.4. Urine sample preparation and SERS measurements

The urine sample aliquots were unfrozen for approximately 30 min at room temperature, and the individual urine samples were homogenized before acquiring their Raman spectra. For the SERS measurements, 150 µL of urine was mixed with 150 µL of AuNP solution. Then, 200 µL of the mixture (urine-AuNPs) was centrifuged at 2000 rpm for 5 min to generate a pellet at the bottom of the centrifuge tube. The generated pellet was dissolved in 50 µL of Type 1 water, and 5 µL of the urine-AuNP mixture was deposited on a copper plate, after which the Raman spectra were measured with a laser intensity of 785 nm, a time of integration of 1000 s, a laser power of 100 mW, and 15 scans. Raman spectra were also acquired for the AuNP solution and urine samples without AuNPs.

2.5. PLS-DA model

Chemometric analysis was performed by using MATLAB® 8.6.0.267246 (R2015b; Math Works Inc. Natick, USA) and PLS Toolbox 8.1 (Eigenvector Research, Inc., Wenatchee, WA, USA). PLS-DA is a multivariate dimensionality reduction tool that is used for feature selection and the classification of data sets containing samples from diverse groups or classes. In this study, the spectral data of different samples were preprocessed using the 2nd derivative (2ndD; order: 2; window: 15 pt; tails: polyinterp)²⁶, which effectively highlights the maximums and minima of the spectral variables within the Raman spectra of urine samples from both healthy subjects and hypertensive patients.

These models were evaluated through the parameters of the confusion matrix, such as sensitivity, specificity, and precision for calibration (Cal), validation (CV), and prediction (Pred)²⁷. In addition, the area under the receiver–operator curve (ROC) was applied, which is a probability curve that displays the performance of a classification model. Furthermore, an analysis of variable importance in projection (VIP) was executed to identify the spectral variables contributing significantly to the discrimination between classes. In this case, a variable is relevant when it has a VIP value > 1.

3.1. Synthesis and characterization of AuNPs

AuNPs were synthesized as per the protocol of Hermanson et al. (2013), albeit with some modifications. A color change of the suspension from light yellow to dark blue to red was observed for monodisperse colloidal AuNPs. The optical property (plasmon resonance) of the AuNP solution was verified with a maximum absorption at 530 nm, corresponding to AuNPs approximately < 50 nm in size (Fig. 1A). In addition, the AuNP particle size distribution was measured by dynamic light scattering (DLS) using Zetasizer LAB equipment. The AuNP particle size distribution was 37 ± 1 nm (Fig. 1B).

3.2. PLS-DA models

Three PLS-DA models were generated; the first PLS-DA model was created with the absorbance data of the visible spectrum of AuNPs. Figure 1A displays the visible spectrum of AuNPs (blue line, NP), urine AH patients with AuNP solution (red line, AH), and urine HV patients with AuNP solution (green line, HV). This figure shows that when the AuNPs are mixed with urine, a new absorption band appears, which may be attributed to the interaction of the AuNPs with the existing metabolites in urine, possibly due to an agglomeration generated in the sample. The sensitivity and specificity for cross-validation for this model were 0.76 and 0.69, respectively; these values indicate acceptable discrimination, suggesting that the concentrations and the type of metabolites in urine differ between HV patients and AH patients. A second model PLS-DA based on the size distribution data was obtained by DLS, as verified by the size of the AuNP clusters with urinary metabolites measured by DLS, giving a sensitivity and specificity of 0.67 and 0.53, respectively.

A third PLS-DA model was generated using the second derivative of SERS spectral data of urine samples from HV and AH patients. The first 6 PLS-DA models presented error values of 0.413, 0.357, 0.299, 0.237, 0.261, and 0.285, and the lowest error value was for model #4, with an error of 0.237 with 4 latent variables, RMSEC of 0.3269 and RMSECV of 0.4275 (Table 1). The model’s performance was evaluated through the parameters of the confusion matrix, such as sensitivity and specificity. Sensitivities and specificities were calculated for the calibration (Cal), validation (Val), and prediction models²⁸. The results indicated a sensitivity of cross-validation (SenCV) of 0.769 (76.9%), sensitivity prediction (SenPred) of 0.864 (86.4%), specificity of cross-validation (SpeCV) of 0.772 (77.2%), specificity prediction of 0.778 (77.8%), accuracy of cross-validation (AccCV) of 0.770 (77.0%), and accuracy prediction of 0.825 (82.5%) (Table 1). The same model was then used to discriminate only patients diagnosed with renal damage and controls, and a new confusion matrix was generated; the sensitivity and specificity were found to be better than for discrimination considering only hypertension (Table 1).

Table 1

PLS-DA classification confusion matrix for hypertensive models.
Hypertensive model
Confusion Table (Cross Validation)
	AH, RAH	HV	N	Sensitivity	Specificity	Accuracy
Predicted as AH, RAH	50	13	65	0,769	0,772	0,770
Predicted as HV	15	44	57
Confusion Table (Prediction)
	AH, RAH	HV	N	Sensitivity	specificity	Accuracy
Predicted as AH, RAH	19	4	22	0,864	0,778	0,825
Predicted as HV	3	14	18
Hypertensive diagnosis with kidney damage model
Confusion Table (Cross Validation)
	RAH	HV	N	Sensitivity	specificity	Accuracy
Predicted as RAH	22	13	26	0,846	0,772	0,795
Predicted as HV	4	44	57
Confusion Table (Prediction)
	RAH	HV	N	Sensitivity	specificity	Accuracy
Predicted as RAH	9	4	9	1,000	0,778	0,852
Predicted as HV	0	14	18

In addition, the ROC curve of the third PLS-DA model was generated to verify how strong the contribution of the spectral variables was with respect to the classification of HV, AH, and RAH patients. The ROC curve was evaluated with the specificity and sensitivity parameters for the validation and prediction model for AH (Fig. 2A) and RAH (Fig. 2B), with an area under the curve of 0.835 (83.5%) for AUC (CV-AH), 0.907 (90.7%) for AUC (Pred-AH), 0.870 (87.0%) for AUC (CV-RAH) and 0.90.1 (90.1%) for AUC (Pred-RAH). Figure 2C and D show the value predicted for AH in the classification of patients into two classes, HV and AH (with the RAH subclass), of the model classification for hypertension. In addition, the contribution of LV scores in the classification of patients displays two well-defined classes of HV and AH patients using a graph of scores with only the first 4 LVs. Notably, even with the obtained sensitivity and specificity, it was necessary to conduct validation to evaluate the clinical efficacy of the classification model for hypertensive patients.

To identify the significant spectral bands for the model, the VIP scores > 1 were identified and compared with the average Raman spectra and the signals generated by the [2ndD] preprocess (Fig. 3). A comparison of the spectrum (data) average of the urine samples from HV with AH was generated (Fig. 3A). Considerable variations were observed in the presence and intensity of spectral peaks in different regions between 400–570, 600–800, 900–1170, and 1200–1400 cm^− 1. When performing second-derivative preprocessing, as shown in Fig. 3B, the different spectral bands were highlighted, with the observation of the highest intensities in the spectral ranges 600–800, 970–1050, and 1300–1350 cm^− 1.

VIP is the measure that indicates the variables that are important and displays their contributions to the model. It helps to select the variables that will be used to develop the predictive model. The calculation of the VIPs of the variables is conducted by means of the weighted sum of the squared correlations between the components and the original variable and whose weightings correspond to the percentage variation explained by the PLS-DA component in the model^29,30.

The VIP scores help identify the most important spectral regions that contribute to the optimal performance of the model³⁰.

In the VIP scores of the Raman spectra matrix of urine samples for the classification model of hypertensive patients, 36 bands were identified that exceeded the limit value, some contributing to the classification model with more weight than others. A total of 36 Raman spectral bands that contribute to the PLS-DA classification model were assigned tentatively, according to the literature, and the bibliographic references of the main Raman bands in urine samples from HV and AH (Table 2).

Table 2

VIP scores matrix spectra Raman urine samples. n.r., not resolvable
Signal	Value VIP scores	Urine peaks (cm^− 1)	Possible assignment
1	1.68	302	n.r.
2	1.41	326	n.r.
3	1.32	432	n.r
4	1.77	531	S–S stretching protein¹¹
5	1.21	607	Creatinine, glycerol^31,32
6	1.79	630	Glycerol, C–S gauche amino acid methionine³²
7	1.25	642	Uric acid, C–C twisting mode of tyrosine^12,32
8	2.07	660	C–S stretching mode of cystine collagen type II^32,14
9	1.91	706	N–H Uric acid, C–S trans amino acid methionine^13,32
10	2.35	726	DNA/RNA bases, hypoxanthine, C–S protein, CH2 rocking adenine^12,32
11	1.24	770	Alanina¹⁶
12	1.97	786	Ring vibration cytosine, DNA O–P–O uracil, thymine^13,32
13	1.31	801	Backbone geometry and phosphate ion interaction³²
14	1.33	828	Glutathione, tyrosine PO₂ stretch DNA phosphodiester, O–P–O stretching DNA/RNA^12,32
15	1.49	894	Phosphodiester deoxyribose³²
16	1.52	907	Creatinine, creatine, hydroxibutyrate³¹
17	1.34	935	C–C stretching mode of proline and valine and protein backbone³¹
18	2.16	993	Ring vibration uric acid, phenylalanine¹³
19	6.53	1008	N–C–N stretching urea, phenylalanine^31,32
20	2.23	1026	C–H stretching phenylalanine¹¹
21	2.85	1037	n.r.
22	1.32	1064	Skeletal C–C stretch of lipids ³²
23	1.25	1089	Po2 stretch, phosphate, histidine, nucleic acid^12,13,32,33
24	1.47	1100	C–C vibration mode of the gauche-bonded chain, amide III³²
25	1.97	1121	C–N stretch protein backbone, vibrations C–O, C–C, C–N uric acid^34,13
26	1.76	1146	C–C lipids, fatty acid^33,32
27	1.68	1271	Amide III band in protein, amide III C–N stretch, C = C fatty acid, typical phospholipids³²
28	1.35	1292	Interring stretching, cytosine³²
29	1.47	1306	CH3/CH2 twisting or bending mode of lipid/collegen³²
30	2.40	1327	CH3CH2 wagging mode in purine bases of nucleic acids³²
31	3.25	1342	CH3, CH2 twisting nucleic acid, wagging protein, G(DNA/RNA), CH deformation protein and carbohydrates^11,34,32
32	1.32	1378	Ring breathing modes DNA, paraffin¹³³²
33	1.62	1404	CH deformation³²
34	1.71	1417	C = C stretching in quinoid ring³²
35	1.17	1428	CH2 creatinine, valine^12,14
36	1.20	1479	Amide III³²

Of the assignments to the VIP scores of the 36 identified bands that exceeded the threshold, 19 bands had an intensity of > 1.5, and 8 bands had the highest intensity of > 2.0 (660, 993, 1007, 1025, 1036, 1327, and 1342 cm^− 1).

Hypertension is a silent disease that is associated with other comorbidities, accounting for 8.5 million deaths worldwide⁸, with a national prevalence in Colombia of 9.08% per 100 inhabitants³⁵. The effect of hypertension on the clinical implications of kidney disease and chronic diseases makes it critical to search for tools that allow early and easy classification of patients with high blood pressure to prevent potential hypertensive kidney damage. This study aimed to develop a noninvasive, rapid, and sensitive method for the diagnosis of AH based on urine analysis using AuNPs and Raman spectroscopy. Enhanced Raman spectroscopy can be used to classify hypertensive and nonhypertensive patients quickly, simply, and economically. We hypothesized that the urine of hypertensive patients presents biochemical changes that can be detected by using AuNPs as Raman signal enhancement agents and by using PLS-DA as a multivariate classification method.

This study is relevant in its endeavor to develop alternatives to the conventional technique of blood pressure measurement for the diagnosis of a condition due to AH, which is a chronic disease that affects millions of people across the world and involves a high risk of developing cardiovascular, renal, and cerebrovascular diseases. In this study, we innovated the combined use of AuNPs and Raman spectroscopy as a photonic technique that facilitates the provision of chemical and structural information from biological samples without any prior preparation or the use of chemical reagents. AuNPs act as Raman signal enhancement agents by inducing an effect known as SERS, which increases the intensity of the spectrum by up to five orders of magnitude, thereby allowing detection of the molecules even at trace levels.

Our study used enhanced Raman spectroscopy to distinguish healthy patients from hypertensive patients and to detect the differences between the molecular composition of the urine of these patients. Moreover, the Raman spectra and PLS-DA allowed the classification of urine samples of the two classes under the study owing to the exclusive spectral characteristics of these samples that facilitated molecular assignments possible to the spectroscopic signals (Table 2). As seen in this table, several bands displayed highly significant scores that contributed significantly to the model, without leaving aside the bands that presented lower scores as they could be significant. This analysis was not based on a single factor or molecular pattern but rather on a combination of several factors, and the relevant information could be extracted from the Raman spectra through PLS-DA, which makes it possible to demonstrate the difference between HV and AH urine spectra and attribute the variations in the molecular patterns.

In the application of the PLS-DA model to classify hypertensive patients and healthy subjects, it was found that in a group of patients who were detected positive for hypertension with the conventional diagnostic tests, 86.4% gave positive results with this model, and in the group of patients who gave negative results with the conventional technique, 77.8% were deemed negative with the PLS-DA model. The present results for both specificity and sensitivity displayed values suitable as a proof of concept that promotes further larger-scale studies toward developing more robust models that guarantee accurate diagnosis. In addition, the present PLS-DA model could identify hypertensive patients with kidney damage, which propels the investigation of a new line of study. It is important to identify hypertensive patients with kidney damage, as they have a higher risk of progressing to chronic kidney failure or developing cardiovascular complications, which may necessitate renal replacement therapy.

In addition, in a clinical setting, the proposed method can be applied for the management of hypertensive patients with kidney damage, such as the possibility of adjusting antihypertensive treatment by preventing or delaying the progression of kidney damage, monitoring response to treatment, or evaluating long-term prognosis. In the future, the proposed method should be optimized, such as with the use of other urinary biomarkers that may be related to hypertensive kidney damage, the use of other metal nanoparticles that may have a different SERS effect, or the use of other statistical techniques that may optimize the classification model.

AH: Hypertensive; HV: controls without a diagnosis of hypertension; RAH: Hypertensive with a diagnosis of kidney disease; VIP: Variable Importance in Projection; PLS-DA: Partial Least Squares-Discriminant Analysis; DBP: Diastolic Blood Pressure, SERS: Surface-Enhanced Raman Spectroscopy; KD: Kidney Disease; CVD: Cardiovascular Disease; DM: Diabetes Mellitus; AuNPs: Gold Nanoparticles; Cal: Calibration; CV: Cross Validation; Pred: Prediction; ROC: Receiver Operating Characteristic Curve; DLS: Dynamic Light Scattering; 2ndD: Second Derivative; LV: Latent Variables; SenCal; Sensitivity Calibration; SenCV: Sensitivity Cross Validation; SenPred: Sensitivity Prediction; SpeCV: Specificity Cross Validation; SpePred: Specificity Prediction; AccCal: Accuracy Calibration; AccCal: Accuracy Calibration; AccPred: Accuracy Prediction

Statement of data availability

All data sets generated for this study are included in the manuscript/supplementary files.

Statement of Compliance with Guidelines and Regulations

We would like to confirm that all procedures and methodologies used in this study were carried out in strict compliance with the relevant guidelines and regulations as stipulated by the journal's editorial policy. Additionally, this study has been approved by the Simon Bolivar University ethics committee (PRO-CEI-USB-0425-00), ensuring the integrity and ethics of our research. We have taken all necessary precautions to ensure that the methods employed are consistent with established standards and to guarantee the validity and reproducibility of our results. We have obtained all required permissions and have ensured that our methods are transparent, ethical, and rigorous.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationship that could be interpreted as a potential conflict of interest.

Author contributions

ENQ, LCPL, and ACEG conceived and designed the project. ENQ, ACEG, SPHR, NJGF, FEA, GAM, LGE and LCPL prepared nanoparticles and took samples. ACEG, LGE performed the experiments. ACEG and LCPL carried out the analysis of the results and their interpretation. ACEG, MML, LCPL, LPL, SPHR, ENQ, and NJGF wrote the manuscript. The draft was reviewed by LCPL, ENQ, NJGF, FEA and MML, and all authors contributed to the article and approved the submitted version.

Funding

We declare that the funds or sources of support received in this specific internal report study were from the Universidad Simón Bolívar, Clinica de la Costa and that the external funding was from the Ministry of Science, Technology and Innovation of Colombia; subsidy: 125380763038; subsidy: 125384467468 and Contract: RC No. 852 de 2019. We clarified that the funder had no role in the design of the study, in the collection and analysis of data, in the decision to publish, or in the preparation of the manuscript.

This work was supported by the Universidad Simón Bolívar and Clínica de la Costa.

Williams, B. et al. ESC/ESH Guidelines for themanagement of arterial hypertension. Eur Heart J 39, 3021–3104 (2018).
Setters, B. & Holmes, H. M. Hypertension in the older adult. Prim Care 44, 529–539 (2017).
O’Shea, P. M., Griffin, T. P. & Fitzgibbon, M. Hypertension: The role of biochemistry in the diagnosis and management. Clin Chim Acta 465, 131–143 (2017).
Lopera Medina, M. M. La enfermedad renal crónica en Colombia: Necesidades en salud y respuesta del Sistema General de Seguridad Social en Salud. Rev Gerencia Pol Salud 15, 212–233 (2016).
Crowley, S. D. & Coffman, T. M. The inextricable role of the kidney in hypertension. J Clin Invest 124, 2341–2347 (2014).
Hamrahian, S. M. & Falkner, B. Hypertension in chronic kidney disease. Adv Exp Med Biol 956, 307–325 (2017).
Tousoulis, D. Arterial hypertension: New concepts in diagnosis and treatment? Hellenic J Cardiol 61, 145–147 (2020).
Zhou, B. et al Worldwide trends in hypertension prevalence and progress in treatment and control from 1990 to 2019: A pooled analysis of 1201 population-representative studies with 104 million participants. Lancet 398, 957–980 (2021).
Cutshaw, G. et al The emerging role of Raman spectroscopy as an omics approach for metabolic profiling and biomarker detection toward precision medicine. Chem Rev 123, 8297–8346 (2023).
Breast Cancer Diagnosis by Surface-Enhanced Raman Scattering (SERS) of Urine _ Enhanced Reader.
Ma, Y. et al Therapeutic prognosis of prostate cancer using surface-enhanced Raman scattering of patient urine and multivariate statistical analysis. J Biophotonics 14, e202000275 (2021).
Lin, J. et al Rapid and label-free urine test based on surface-enhanced Raman spectroscopy for the noninvasive detection of colorectal cancer at different stages. Biomed Opt Express 11, 7109–7119 (2020).
Pappu, R., Prakasarao, A., Dornadula, K. & Singaravelu, G. Raman spectroscopic characterization of urine of normal and cervical cancer subjects. in. SPIE Proceedings (SPIE. 2017) 10054, 22–28 (2017).
Flores-Guerrero, J. L. et al Novel assessment of urinary albumin excretion in type 2 diabetes patients by Raman spectroscopy. Diagnostics (Basel) 10, 141 (2020).
de Souza Vieira, E. E., Bispo, J. A. M., Silveira, L. & Fernandes, A. B. Discrimination model applied to urinalysis of patients with diabetes and hypertension aiming at diagnosis of chronic kidney disease by Raman spectroscopy. Lasers Med Sci 32, 1605–1613 (2017).
Chen, C. et al. Urine Raman spectroscopy for rapid and inexpensive diagnosis of chronic renal failure (CRF) using multiple classification algorithms. Optik 203, 164043 (2020).
Aitekenov, S. et al. Surface-enhanced Raman spectroscopy (SERS) for protein determination in human urine. Sens Biosensing Res 38, 100535 (2022).
Nikelshparg, E. I. et al Detection of hypertension-induced changes in erythrocytes by SERS nanosensors. Biosensors (Basel) 12, 32 (2022).
de Sousa Vieira, E. E. et al. Biochemical analysis of urine samples from diabetic and hypertensive patients without renal dysfunction using spectrophotometry and Raman spectroscopy techniques aiming classification and diagnosis. Bioengineering (Basel) 9, 500 (2022).
Yang, H. et al Noninvasive and prospective diagnosis of coronary heart disease with urine using surface-enhanced Raman spectroscopy. Analyst 143, 2235–2242 (2018).
Kaman, P., Dutta, P. & Bhattacharyya, A. Synthesis of Gold Nanoparticles from Metarhizium anisopliae for Management of Blast Disease of Rice and Its Effect on Soil Biological Index and Physicochemical Properties 14, 1136233 (2022).
Narband, N. et al The interaction between gold nanoparticles and cationic and anionic dyes: Enhanced UV–visible absorption. Phys Chem Chem Phys 11, 10513–10518 (2009).
Huang, H. & Yang, X. Synthesis of chitosan-stabilized gold nanoparticles in the absence/presence of tripolyphosphate. Biomacromolecules 5, 2340–2346 (2004).
Sutriyo, et al Synthesis of gold nanoparticles with polyamidoamine (Pamam) Generation 4 dendrimer as stabilizing agent for CT scan contrast agent. in Macromol Symp (Wiley-VCH Verlag. 2015) 353, 96–101 (2015).
Wise, B. M., Shaver, J. M., Gallagher, N. B., Rasmus, W. W. & Koch, B. R. S. Continuum Parameter Number of LVs PRESS Chemometrics Tutorial for PLS_Toolbox and Solo. at http://www.eigenvector.com.
Ruiz-Perez, D., Guan, H., Madhivanan, P., Mathee, K. & Narasimhan, G. So you think you can PLS-DA? BMC Bioinformatics 21, 1–10 (2020).
de Araújo Gomes, A., Azcarate, S. M., Diniz, P. H. G. D., de Sousa Fernandes, D. D. & Veras, G. Variable selection in the chemometric treatment of food data: A tutorial review. Food Chem 370, 131072 (2022).
Surmacki, J. M., Woodhams, B. J., Haslehurst, A., Ponder, B. A. J. & Bohndiek, S. E. Raman microspectroscopy for accurate identification of primary human bronchial epithelial cells. Sci Rep 8, 12604 (2018).
Banerjee, P. et al Identification of key contributory factors responsible for vascular dysfunction in idiopathic recurrent spontaneous miscarriage. PLOS ONE 8, e80940 (2013).
Moreira, L. P. et al Raman spectroscopy applied to identify metabolites in urine of physically active subjects. J Photochem Photobiol B 176, 92–99 (2017).
Raman Spectroscopy of Biological Tissues _ Enhanced Reader.
Primera-Pedrozo, O. M. et al Nanotechnology-based detection of explosives and biological agents simulants. IEEE Sens J 8, 963–973 (2008).
Li, J. et al Raman spectroscopy as a diagnostic tool for monitoring acute nephritis. J Biophotonics 9, 260–269 (2016).
Situación de la enfermedad renal crónica, la hipertensión arterial y diabetes mellitus en Colombia. (2020). Cuenta de Alto Costo. at https://cuentadealtocosto.org/site/publicaciones/situacion-de-la-enfermedad-renal-cronica-la-hipertension-arterial-y-diabetes-mellitus-en-colombia-2020/?1665593257100.
35. Situación de la enfermedad renal crónica, la hipertensión arterial y diabetes mellitus en Colombia. (2020). Cuenta de Alto Costo. at https://cuentadealtocosto.org/site/publicaciones/situacion-de-la-enfermedad-renal-cronica-la-hipertension-arterial-y-diabetes-mellitus-en-colombia-2020/?1665593257100.

No competing interests reported.

supplementaryfiles.xlsx

Download PDF

Journal Publication

published 06 Feb, 2024

Read the published version in Scientific Reports →

Editorial decision: Revision requested
05 Nov, 2023
Reviews received at journal
28 Oct, 2023
Reviewers agreed at journal
22 Oct, 2023
Reviewers invited by journal
08 Sep, 2023
Editor assigned by journal
08 Sep, 2023
Editor invited by journal
04 Sep, 2023
Submission checks completed at journal
04 Sep, 2023
First submitted to journal
28 Aug, 2023

You are reading this latest preprint version

Surface-Enhanced Raman Spectroscopy in Urinalysis of Hypertension Patients with Kidney Disease

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Materials and methods

2.1. Materials

2.2. Patients, sample collection, and storage

2.3. Synthesis and characterization of AuNPs

2.4. Urine sample preparation and SERS measurements

2.5. PLS-DA model

3. Results

3.1. Synthesis and characterization of AuNPs

3.2. PLS-DA models

4. Discussion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1