Enantioanalysis of Leucine in Whole Blood Samples Using Enantioselective, Stochastic Sensors

Abstract

Enantioanalysis of amino acids became a key factor in the metabolomics of cancer. As a screening method, it can provide information about the state of health of patients. The main purpose of the study is to develop a highly reliable enantioanalysis method for the determination of D-, and L-leucine in biological samples in order to establish their role as biomarkers in the diagnosis of breast cancer. Two enantioselective stochastic sensors based on N-methyl-fullero-pyrrolidine in graphite and graphene nanopowder pastes were designed, characterized, and validated for the enantioanalysis of leucine in whole blood. Different signatures were recorded for the biomarkers when the stochastic sensors were used, proving their enantioselectivity. In addition, limits for detection on the order of ag L⁻¹ were recorded for each of the enantiomers of leucine when the proposed enantioselective stochastic sensors were used. The wide linear concentration ranges facilitated the assay of the L-leucine in healthy volunteers, and also in patients confirmed with breast cancer. Recoveries of one enantiomer in the presence of the other enantiomer in whole blood samples, higher than 96.50%, proved that the enantioanalysis of enantiomers can be performed reliably from whole blood samples.

1. Introduction

Amino acids metabolism is of high importance for cell proliferation, as well as for personalized treatment and early diagnosis of cancer [1,2,3,4]. Amino acids have an important role in the redox process, energy regulation, biosynthesis, and maintenance of homeostasis [5]. They usually feed cancer cells by providing building blocks for cancer cell growth [5]. The major function of amino acids in mammalian cancer cells is as substrates for new protein synthesis [6].

Racemization of amino acids in the body, while cancer cells are forming, was shown for different amino acids, such as aspartic acid [7]. In addition, distortion of DNA can produce the D—enantiomer of an amino acid. While L-amino acids are the most common in healthy people, the enantioanalysis of different biological samples showed the presence of D-amino acids in patients confirmed with different types of cancer [2,3,4,8].

The first evidence for the presence of D-amino acids in cancer was given in 1939 by Kögl and Erxleben [9], when they reported that some D-amino acids including D-leucine were found in tumor proteins and that the development of the tumor was facilitated by the presence of these amino acids, including D-leucine. While initially the assay of a chiral compound was done despite its stereochemistry, the evidence that differentiation between enantiomers may conducive for the diagnosis of different diseases. Discovered in 1819 in its impure form, leucine is an amino acid with high implications in carcinogenesis [10]. In 1820, the crystalline form of leucine was extracted from muscles and wool [10]. The name leucine is given by the Greek word leukos, which has the meaning “white”; the motivation of the name is given by the fact that the purification of leucine extracted from natural products provided a white, crystalline state [10]. Leucine is an essential amino acid due to its proven biochemical properties: it contributes to protein synthesis and is involved in metabolomics including cancer metabolomics [10].

Leucine is well known for its involvement in controlling whole blood sugar levels and improving the growth/recovery of muscles and bone tissues. Leucine is also responsible for the synthesis of growth hormones. Leucine also has a role in the prevention of breaking muscle proteins due to injuries like stress [10]. Leucine is part of the branched-chain amino acid valine, isoleucine, and leucine. It is one of the two (the other being lysine) ketogenic amino acids.

Acetyl coenzyme A and acetoacetate are the main metabolic end products produced during leucine metabolism [11]. Leucine and β-hydroxy β-methylbutyric acid (one of the leucine metabolites), have shown a great role in the human body. They induce biosynthesis of proteins by phosphorylation reactions. The metabolism of the dietary leucine takes place in the liver, adipose tissue, and muscle tissue, with the synthesis of sterols and derivatives being a result of the metabolism of leucine in the adipose and muscle tissues [11]. The pathway of L-leucine in the body [11] shows that the branched-chain amino acid aminotransferase enzyme is responsible for the initial transformation of L-leucine into α-ketoisocaproate (known as α-KIC) [11]. Next, mitochondrial enzyme branched-chain α-ketoacid dehydrogenase converts α-KIC into isovaleryl-coenzyme A, which is further transformed using the isovaleryl-coenzyme A dehydrogenase into methylcrotonyl-coenzyme A [11]. L-leucine also contributes to the increase of biotin during deficiency, when the hydroxymethylbutyrate (HMB) can be synthesized from methylcrotonyl-coenzyme A using the enoyl-coenzyme A hydratase as well as an unknown thioesterase enzyme. The chain of reactions involved in the conversion of methylcrotonyl-coenzyme A into hydroxymethylbutyrate-coenzyme A ends with the final conversion into hydroxymethylbutyrate. In the liver an extremely small amount of α-KIC is metabolized using the cytosolic enzyme 4-hydroxyphenylpyruvate dioxygenase (KIC dioxygenase); in addition, in the liver, the main end product is the hydroxymethylbutyrate. The described chain of transformations—starting from L-leucine to α-KIC and then to HMB—is specific to healthy people [11]. Leucine aminomutase transforms a very small amount of L-leucine into β-leucine, which is transformed further into β-ketoisocaproate (β-KIC), β-ketoisocaproyl-coenzyme A, and acetyl-coenzyme A. HMB is transformed into β-hydroxy β-methylbutyryl-coenzyme A which undergoes transformation using the enoyl-coenzyme A hydratase with the main product β-methylcrotonyl-coenzyme A and hydroxymethylglutaryl-coenzyme A [11].

This article proposed the enantioanalysis of leucine in whole blood samples of patients, confirmed with breast cancer versus healthy volunteers. Leucine is an essential amino acid belonging to the branched-chain amino acids. The role of leucine in breast cancer biochemistry and treatment was studied in different occasions [12,13,14,15,16,17]. Xiao et al. [12] proved that leucine deprivation inhibits proliferation and induces apoptosis of human breast cancer cells. Troup et al. [13] showed that reduced expression of leucine is associated with poor outcomes in node-negative invasive breast cancer. Shennan et al. [14] studied the L-leucine transport in human breast cancer cells. The role of leucine as a zipper [15,16] was shown in studies by Lin et al. [15] and Jeong et al. [16]. The role of D-leucine as a new biomarker for breast cancer was also investigated.

Due to the complexity of the matrix of whole blood, and the low levels of leucine enantiomers in whole blood, the stochastic sensors were selected as new screening tools. Stochastic sensors are known for their reliability recorded in the biomedical analysis of different biological samples (whole blood, saliva, urine, tumoral tissue, cerebrospinal fluid) when amino acids [2,3,4] or proteins [17,18,19] were analyzed. They are able to perform qualitative analysis—for identification of the enantiomer [2,3,4,20,21,22,23,24]—when the signature of the enantiomer (t_off value) is used, and also a quantitative analysis by using the t_on value measured in between two signatures (t_off values) (Scheme 1); they are also able to identify the stereochemistry of a compound, and also, they are able to distinguish the amino acids from a DNA chain.

The working principle of the stochastic sensors is based on current development. The principle of current development is based on channel conductivity; in the first stage, the enantiomer is getting into the channel and blocking it while the current intensity is dropping to zero (the signature of the enantiomer given by the t_off value is characterized in this step); inside the channel, binding and redox processes take place, and equilibrium is achieved (t_on value, used for the quantitative assay of the enantiomers, is characterized in this step).

While the first stage is used for identification of the enantiomer, the second stage is used for the determination of the concentration of each enantiomer.

Utilization of the stochastic sensors for biomedical analysis is based on the advantages of the stochastic sensors versus classical electrochemical sensors, which are that the signature of the enantiomers does not depend on the composition of the matrix, but only on the size and geometry of the enantiomer. In addition, the quantitative step takes place inside the channel, and therefore is less influenced by the composition of the sample from where the enantiomer is determined. Graphite and graphene matrices are well known for sensor design due to their conductivity and high reliability when they are used as a paste in the sensor’s design [25]. These matrices are able to keep the channel in the best shape needed for obtaining the stochastic signal. The fullerene derivative N-methyl-fullero-pyrrolidine was selected due to its capacity to provide the necessary channels for stochastic sensing. Fe₂O₃ was added to the graphite paste in order to improve its conductivity, which facilitated, at the selected potential (125 mV), signatures and values for t_on on the order of seconds, and therefore, the reading was done with high reliability.

To date, for the assay of leucine, there was proposed a glassy carbon electrode modified with multiwall carbon nanotubes (presenting a limit of detection of 9 µmol L⁻¹) [26], and high-performance liquid chromatography [27]. The enantioanalysis of leucine was done to date using a low-cost high-performance liquid chromatographic method [28] as well as amperometric biosensors based on diamond paste [29] able to provide limits of detection on the order of pmol L⁻¹.

The applications of electrochemical sensors in biomedical analysis are guaranteed by the following advantages versus other methods of analysis, such as the enzyme-linked immunosorbent assay, high-performance liquid chromatography, electrophoresis, gas chromatography, and spectrometric methods of analysis: non-sampling, or a minimum sampling are required before any measurement is performed; wide working concentration ranges are available; the limits of detection are very low; the selectivity is quite high for the amperometric and potentiometric sensors, and can be improved by using a biological substance such as an enzyme or antibody; the sensitivity is also very high; they are cost-effective as such and also regarding the instrumentation needed to take up the signal and process it; and the time of analysis is far shorter than for any other analytical technique (analysis are usually performed within minutes or even seconds) [30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48].

The design of the sensors used in biomedical analysis influenced the response characteristics of the sensors (e.g., their sensitivity), and also had a role in improving the selectivity of determining the selected biomarkers in highly complex matrices such as biological samples. The utilization of nanomaterials for the design of electrochemical sensors proved to have a high role in the reliable assays of the amino acids in the biological samples [30]. Utilization of molecular imprinted polymers [29] is trending due to the reliable design of sensors, and the possibility of continuous in vivo monitoring of specific biomarkers. Hydrogels were used previously in the design of multifunctional platforms, also offering the possibility of high motion of the biomolecules to the detection side [37]. Utilization of nanorods proved to improve the response characteristics of the electrochemical sensors used for biomedical analysis [47]. Miniaturization of the detection cell to lab-on-a-chip type of design [30,33,37] favored easy transportation and handling of the system, especially for on-site type of measurements. In addition, to avoid cross-contamination, disposable sensors were developed for biomedical analysis [45]. These sensors can be easily be used for point-of-care diagnostics, where fast detection and minimum (preferably no sample preparation) sample preparation is needed [32,35,40]. Smartphones became preferable tools for signal acquisition, processing, and for sending the results to medical doctors and specialized databases and laboratories [35,36]. Commercialization of sensors for biomedical analysis needs smart, portable, and compact detection systems, on which the sensors should easily fit, connected to smartphones [38].

The novelty of the paper is given by the utilization of N-methyl-fullero-pyrrolidine as a modifier of graphite/Fe₂O₃ and of nanographene pastes, for the design of enantioselective stochastic sensors and their utilization for enantioanalysis of leucine in whole blood samples, also by the stochastic method capacity to perform an enantioselective molecular recognition of the enantiomers followed by their quantification, which may facilitate a highly reliable diagnosis of breast cancer. While most of the paper shows the analysis of leucine, and not its enantioanalysis, the main purpose of the study is to develop a highly reliable enantioanalysis method for the determination of D-, and L-leucine in biological samples in order to establish their role as biomarkers in the diagnosis of breast cancer.

2. Materials and Methods

L- and D-leucine, graphite, Fe₂O₃, graphene nanopowder, N-methyl-fullero-pyrrolidine, and paraffin oil were purchased from Sigma Aldrich. All solutions of L- and D-leucine were prepared using phosphate buffer of pH 7.40. The concentration range used for both enantiomers was between 1 × 10⁻²⁰ and 1 × 10⁻² mol L⁻¹.

The design of the enantioselective stochastic sensor was done as follows: 100 mg of graphite/graphene nanopowder were physically mixed with 20 mg of N-methyl-fullero-pyrrolidine. To the graphite powder mixture there were added 10 mg Fe₂O₃. To each of the powders, 30µL of paraffin oil was added to form a homogeneous paste. Each of the pastes was placed in a nonconducting 3D plastic tube printed in our laboratory using a 3D printer. The inner diameter of each tube was 25 µm, and the length was 1 cm. A Ag wire was used to connect the paste with the external circuit.

The morphology of the designed pastes was investigated using scanning electron microscopy (SEM) (Inspect S, FEI Company, Eindhoven, The Netherlands). In order to obtain a good resolution of the microscopy images, the pastes were analyzed using the LFD detector (low vacuum), at a high voltage (HV) of 30 kV, and magnification of 1600×.

The morphology of the active surface of the stochastic sensors is shown in Figure 1.

In Figure 1, agglomerations of particles and channels in asymmetric formations can be observed.

The stochastic method is based on the utilization of the chronoamperometric technique. A potential of 125 mV versus Ag/AgCl was applied. The AUTOLAB/PGSTAT 12 Potentiostat/Galvanostat (Metrohm, The Netherlands) was used for all measurements. The electrochemical cell comprised the enantioselective stochastic sensor (as a working sensor), an Ag/AgCl reference electrode, and a Pt wire as an auxiliary electrode. The electrochemical measurements were performed at 125 mV vs. Ag/AgCl, and at 25 °C. Diagrams like those shown in Figure 2 were obtained when the whole blood was screened. A standard solution of enantiomers as well as whole blood samples spiked with standard solutions of enantiomers were used for the calibration of the enantioselective stochastic sensors; no differences between the response characteristics of the sensors were recorded when the calibration was done for the standard solution or in whole blood samples. Based on the signatures of the enantiomers of leucine (values of t_off), the enantiomers were identified in the diagrams recorded for whole blood samples. In between two signatures, the t_on was read and used for the determination of response characteristics of the sensors; linear regression method was used to determine a, and b parameters of the calibration equation in the form of 1/t_on = a + b × C_enantiomer. The unknown concentration of the enantiomer was determined using the calibration equation.

The whole blood samples obtained from confirmed patients with breast cancer as well as whole blood samples from healthy volunteers were provided by the University Hospital Bucharest under the protocol approved by the University of Medicine and Pharmacy “Carol Davila”, from Bucharest (ethics committee approval no 75/2015). No pretreatment of the samples was required before analysis. The samples were placed into an electrochemical cell; the diagram was recorded. First of all, the molecular recognition of L- and D-leucine was performed by identifying their signatures (t_off values) in the diagrams. Their concentrations were determined according to the stochastic method described above.

3. Results and Discussions

3.1. Response Characteristics

The stochastic mode was used for the determination of the response characteristics of the enantioselective stochastic sensors. The response characteristics are given in

Table 1. Characteristics of response for the enantioselective stochastic sensors used for the enantioanalysis of leucine.

Stochastic Sensor Based on N-Methyl-fullero-pyrrolidine and	Leucine	Signature toff (s)	Equation of Calibration 1, R	Sensitivity, s−1 g−1 L	Limit of Determination, ag L−1	Linear Concentration Range, g L−1
Graphite/Fe2O3	L	0.6	1/ton = 0.04 + 8.17 × 1013C r = 0.9993	8.17 × 1013	10.00	1 × 10−17–1 × 10−5
	D	0.8	1/ton = 0.01 + 1.42 × 1013C r = 0.9994	1.42 × 1013	100.00	1 × 10−16–1 × 10−9
Nanographene	L	2.2	1/ton = 0.02 + 3.44 × 1012C r = 0.9995	3.44 × 1012	100.00	1 × 10−16–1 × 10−4
	D	0.9	1/ton = 0.02 + 1.35 × 1015C r = 0.9996	1.35 × 1015	1.00	1 × 10−18–1 × 10−6

¹ <C> = mol L⁻¹; <t_on> = s.

.

All response characteristics reported in Table 1 were obtained by applying a potential of 125 mV versus Ag/AgCl (as reference electrode), at a constant temperature of 25 °C.

The signatures of the two enantiomers are different when the same stochastic sensor was used, proving that the sensors are enantioselective, and they can be used for the simultaneous assay of the two enantiomers in biological samples. Linear concentration ranges are wide: twelve decades of concentration for L-leucine, seven decades of concentration for the D-leucine when the sensor based on graphite/Fe₂O₃ was used, and twelve decades of concentration for D-leucine when the sensor based on nanographene was used, allowing the enantioanalysis of leucine in whole blood samples despite people’s state of health. Very low limits of detection (determined as the lowest value of concentration from the linear concentration range)—all in the order of ag L⁻¹—were recorded; for the assay of L-leucine the lowest limit of detection was recorded when the enantioselective stochastic sensor based on graphite/Fe₂O₃ was used, while for the assay of D-leucine the lowest limit of detection was recorded for the enantioselective stochastic sensor based on nanographene. L-leucine was determined with the highest sensitivity when the sensor based on graphite/Fe₂O₃ was used, while D-leucine was determined with the highest sensitivity when the sensor based on nanographene was used.

3.2. Selectivity of the Enantioselective Stochastic Sensors

The selectivity of the stochastic sensors is given by the difference between the signatures (t_off values) recorded for the enantiomers of leucine and those obtained for CA15-3, CEA, HER2, p53, Ki67, maspin, CA19-9, and L- and D-serine—the usual biomarkers used for the establishment of the diagnostic of breast cancer.

Results shown in

Table 2. Selectivity of the stochastic sensors.

Stochastic Sensor Based on N-Methyl-fullero-pyrolidine and	CA15-3	CEA	HER2	Maspin	Ki67	CA19-9	p53	L-Leucine	D-Leucine	L-Serine	D-Serine
	Signature (s)
Graphite/Fe2O3	1.1	1.5	2.2	1.9	3.0	2.4	3.5	0.6	0.8	0.2	1.7
Nanographene	0.2	0.6	3.0	2.5	3.2	2.8	1.7	2.2	0.9	1.3	2.0

prove that none of the other biomarkers interfere in the enantioanalysis of leucine despite the matrix used for the stochastic sensor’s design.

3.3. Stability and Reproducibility Measurements

Ten enantioselective stochastic sensors from each of the two types designed for the enantioanalysis of leucine were constructed. Response characteristics were measured for one month on a daily basis. The variation in sensitivities of the proposed sensors was lower than 0.15%; accordingly, the design of the proposed stochastic sensors is reproducible. In addition, variations of the sensitivities were lower than 1.20% for measurements made during the one month, proving the one-month stability of the stochastic sensors.

3.4. Enantioanalysis of Leucine in Whole Blood Samples

The whole blood samples collected from the patients were used without any processing. The measurements were performed accordingly with the description of the stochastic method from above. After the diagrams were obtained (Figure 2), the signatures of the L- and D-leucine were identified. For each of the L- and D-enantiomers, readings of t_on values (in between two consecutive t_off values were done) were performed for the determination of the concentrations of L- and D-leucine. The t_on values were introduced in the equations of calibration (Table 1) for the determination of the concentrations of the enantiomers of leucine. Validation tests were performed for the assay of L- and D-leucine in whole blood samples.

To validate the sensors, synthetic mixtures of enantiomers (in different ratios) were spiked into the whole blood samples, and the recovery of enantiomers was performed (

Table 3. Recovery tests of L- and D-leucine in whole blood samples (N = 10).

							Recovery %
L:D	1:99		1:50		1:25		1:1		25:1		50:1		99:1
Enantiomer	L	D	L	D	L	D	L	D	L	D	L	D	L	D
Graphite/Fe2O3 based sensor	99.10 ± 0.05	96.98 ± 0.02	97.95 ± 0.04	99.18 ± 0.02	99.00 ± 0.02	98.75 ± 0.03	98.82 ± 0.02	98.99 ± 0.01	99.15 ± 0.01	99.10 ± 0.03	99.99 ± 0.01	97.98 ± 0.02	98.16 ± 0.03	99.90 ± 0.02
Nanographene based sensor	99.32 ± 0.02	96.50 ± 0.03	98.00 ± 0.02	99.53 ± 0.03	99.18 ± 0.03	99.65 ± 0.02	98.09 ± 0.01	99.99 ± 0.03	99.13 ± 0.02	99.76 ± 0.02	99.12 ± 0.03	97.00 ± 0.04	97.43 ± 0.02	99.66 ± 0.03

). Determinations of the enantiomers of leucine were performed before and after the addition of the mixtures, in order to calculate the recovery of the known amount added. Different ratios of the enantiomers were selected in order to determine if the ratio of the enantiomers influenced in any way the reliability of the enantioanalysis of leucine (Table 3).

Table 3 shows that high recoveries of the enantiomers were obtained in the whole blood, despite the ratios between L- and D-leucine in the whole blood. This proved the accuracy and high reliability of the measurements performed with the enantioselective stochastic sensors, despite the ratio of which the enantiomers are found in the whole blood sample.

Real whole blood samples collected from patients confirmed with breast cancer and healthy volunteers were analyzed using the enantioselective stochastic sensors. Results obtained are shown in

Table 4. Enantioanalysis of leucine in whole blood samples using enantioselective stochastic sensors (N = 10).

Sample No.	State of Health	L-Leucine, pg mL−1		D-Leucine, ng mL−1
Stochastic Sensor Based on N-Methyl-fullero-pyrrolidine and		Graphite/Fe2O3	Nanographene	Graphite/Fe2O3	Nanographene
1	Confirmed with breast cancer	8.62 ± 0.02	8.08 ± 0.01	3.52 ± 0.02	3.30 ± 0.03
2		0.48 ± 0.01	0.49 ± 0.03	5.00 ± 0.03	4.75 ± 0.02
3		8.25 ± 0.01	8.71 ± 0.03	0.20 ± 0.01	0.18 ± 0.02
4		15.48 ± 0.02	16.02 ± 0.03	1.87 ± 0.02	1.69 ± 0.03
5		6.11 ± 0.01	5.50 ± 0.03	1.00 ± 0.02	1.00 ± 0.03
6		1.24 ± 0.03	0.98 ± 0.02	2.00 ± 0.01	2.17 ± 0.02
7		3.47 ± 0.01	2.93 ± 0.02	2.60 ± 0.03	2.58 ± 0.01
8		0.08 ± 0.01	0.07 ± 0.02	33.10 ± 0.02	35.01 ± 0.01
9		2.91 ± 0.01	2.26 ± 0.03	5.23 ± 0.03	5.00 ± 0.01
10		8.30 ± 0.03	8.12 ± 0.01	1.17 ± 0.03	1.18 ± 0.01
1	Healthy volunteers	18.21 ± 0.01	18.34 ± 0.02	- *	- *
2		20.58 ± 0.02	20.84 ± 0.01	- *	- *
3		5.48 ± 0.01	5.12 ± 0.03	- *	- *
4		32.25 ± 0.02	32.40 ± 0.01	- *	- *
5		7.12 ± 0.01	7.15 ± 0.03	- *	- *
6		27.16 ± 0.02	27.19 ± 0.01	- *	- *
7		52.01 ± 0.03	51.15 ± 0.02	- *	- *
8		4.89 ± 0.01	4.68 ± 0.03	- *	- *
9		51.97 ± 0.03	52.53 ± 0.01	- *	- *
10		43.47 ± 0.01	43.50 ± 0.02	- *	- *
t-test		2.96		3.01

* D-leucine was not found.

.

A paired Student t-test was conducted at a confidence level of 99.90%. At the 99.00% confidence level, the tabulated theoretical value is 4.032. The calculated t-values for each enantiomer of leucine were less than 3.10 (which is less than the tabulated value 4.13), indicating that there is no statistically significant difference between the results obtained using the proposed enantioselective stochastic sensors (Table 4), and that enantioselective stochastic sensors can be relied upon for the molecular identification and quantification of L- and D-leucine in whole blood samples. Furthermore, the D-leucine was only found in the samples collected from confirmed patients with breast cancer, and not in the whole blood from healthy volunteers.

Compared with the sensors proposed earlier for the analysis [24] and enantioanalysis [27] of leucine, the proposed enantioselective stochastic sensors have a far lower limit of detection (on the order of ag L⁻¹). In addition, a wider linear concentration range was recorded for the proposed sensors than for those reported before [24,27]. High selectivity was recorded versus other biomarkers that may be present in the whole blood of patients with breast cancer. High recoveries of one enantiomer in the presence of the other enantiomer as well as the results of paired Student t-tests proved that the enantioselective stochastic sensors can reliably be used for the enantioanalysis of leucine in whole blood samples. Compared with the method proposed earlier by Hormozi Jangi, et al. [49] this screening method provides wider linear concentration ranges, higher sensitivities, and higher accuracy for the enantioanalysis of L- and D-leucine when they are mixed in different ratios.

4. Conclusions

Two enantioselective stochastic sensors were designed, characterized, and validated for the enantioanalysis of leucine in whole blood samples. The utilization of nanographene as a matrix for the sensor’s design favored the decrease of the limit of detection of D-leucine to 1 ag L⁻¹. The sensors are not only enantioselective but also selective versus a series of other biomarkers such as CA15-3, CEA, HER2, maspin, Ki67, CA19-9, and p53, which are the main biomarkers used in the diagnosis of breast cancer. The proposed enantioselective stochastic sensors proved to have great features in biomedical analysis. Their utilization for the screening of whole blood samples can bring the screening test very close to a diagnostic test, because the D-leucine was only found in confirmed patients with breast cancer, and not in healthy volunteers. Accordingly, D-leucine may be considered a biomarker for breast cancer.