PROPOSAL FOR A FILTERLESS FLUORESCENCE SENSOR

FOR SNP GENOTYPING

K. Yamasaki

, H. Nakazawa

1,2

, N. Misawa

1,3

, M. Ishida

1,3

and K. Sawada

1,3,4

Toyohashi University of Technology (TUT), 1-1 Hibarigaoka, Tempaku-cho, Toyohashi, Aichi 441-8580, Japan

JSPS Research Fellow, Chiyoda, Tokyo 102-8472, Japan

Electronics-Inspired Interdisciplinary Research Institute (EIIRIS), Toyohashi, Aichi 441-8580, Japan

JST-CREST, Chiyoda, Tokyo 102-8666, Japan

Keywords: SNP genotyping, Fluorescence, Filterless, Multiwavelength.

Abstract: This study describes a biosensor for single nucleotide polymorphism (SNP) genotyping based on the

filterless fluorescence detection methods. The filterless fluorescence sensor is able to distinguish lights with

more than two different wavelengths without optical filters, mirrors, and gratings. From the final results, we

observed that emission lights form the “fluorescein isothiocyanate (i.e., FITC)” and the “sulforhodamine

101 acid chloride (i.e., Texas Red)”, which are kinds of fluorescent dyes commonly used in SNP genotyping,

were detected with less interference using the filterless fluorescence sensor. Thus, our approach is effective

for SNP genotyping with low cost and high portability.

1 INTRODUCTION

Micro-total analysis systems (μ-TAS) are highly

desirable for the detection of various types of bio-

chemical information with low cost and high porta-

bility. Furthermore, μ-TAS offers the potential for

highly efficient, simultaneous analysis of a large

number of biologically important molecules in ge-

nomic, proteomic and metabolic studies.

μ-TAS compounds pumps, valves, reactors, hea-

ters, micro-fluidic channels, and sensors. The vari-

ous types of sensors that have functionality for bio-

chemical analyses have recently been studied, with

research aimed at compounding with μ-TAS.

One of the most useful sensing methods in bio-

chemistry is the fluorescence detection method. We

have previously devised a filterless fluorescence

sensor which can be applied to the detection of fluo-

rescence (Maruyama et al., 2006); (Maruyama et al.,

2006); (Nakazawa et al., 2011). This sensor detected

several signals with different wavelengths simulta-

neously, without the need for optical filters, mirrors,

and gratings. In this paper, we purpose an ability of

a filterless fluorescence sensor to apply to single

nucleotide polymorphism (SNP) genotyping based

on the filterless fluorescence detection method.

2 FILTERLESS FLUORESCENCE

SENSOR

2.1 Sensor Structure

Figure 1 shows a photomicrograph of the devised

sensor. The devised sensor was fabricated in our

laboratory using 5-μm-rule, N-substrate, 1P1M (1-

poly and 1-metal), and single well modified com-

plementary metal-oxide semiconductor fabrication

technology. The size of the sensing area is 300 × 300

µm

, and the depth of the p-n junction is about 4 µm

beneath the sensing area.

Figure 1: Photomicrograph of the fabricated filterless

fluorescence sensor.

185

Yamasaki K., Nakazawa H., Misawa N., Ishida M. and Sawada K..

PROPOSAL FOR A FILTERLESS FLUORESCENCE SENSOR FOR SNP GENOTYPING.

DOI: 10.5220/0003774201850189

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2012), pages 185-189

ISBN: 978-989-8425-91-1

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

Figure 2 shows the concept and configuration of

the devised sensor. The devised sensor is based on a

photo-gate structure. The sensing area of the device

consists of four layers: n-type poly-Silicon (500

nm)/SiO

(90 nm)/p-well (4 µm)/n-type silicon sub-

strate.

Figure 2: Schematic diagram of section A-B shown in Fig.

1. The cross-section structure is based on a photo-gate

structure.

2.2 Basic Principle

The principle of the filterless fluorescence detection

is based on the variation of optical absorption depth

with wavelength (Mckelvey, 1996). When lights

with two different wavelengths (e.g., fluorescence

and the excitation light) are incident simultaneously,

the currents I generated at absorption depths w

and

are given by the following equations (1):

ሺ

1-e

-α

ሻ

ሺ

1-e

-α

ሻ

ሺ

1-e

-α

ሻ

ሺ

1-e

-α

ሻ

(1)

where φ

and φ

are the intensities at wavelengths λ

and λ

with absorption coefficients α

and α

, respec-

tively, S is the size of the sensing region, q is the

elementary charge, h is the Planck constant, and c is

the speed of light in a vacuum. Both illumination

intensities (φ

and φ

) are obtained by solving these

simultaneous equations.

Figure 3: Potential distribution beneath the sensing area.

The energy band structure beneath the sensing

region shown in Fig. 3 is essential for fluorescence

sensing. A potential well is formed at the surface of

the silicon substrate beneath the sensing region. The

most important consideration for fluorescence sens-

ing with the devised sensor is that the absorption

depth w is tunable, because there is a variation in the

absorption depth w for different wavelengths. In the

devised sensor, the absorption depth w can be con-

trolled by changing the photo-gate voltage, and

electrons generated within the selected absorption

depth w are collected for read out.

2.3 Sensing Procedure

Figure 4 shows a potential diagram for the intensity

measurements of fluorescence and excitation light. If

the photo-gate voltage is set to V

then the absorp-

tion depth changes to w

. When light is incident on

the sensing area, photons are absorbed in the deple-

tion layer below the sensing area. Finally, the cur-

rents, I

, generated at absorption depth w

are read

out via the output node. The currents, I

, generated

at absorption depth w

are similarly obtained by

changing the photo-gate voltage from V

to V

Figure 4: Potential diagram for the intensity measurements

of fluorescence and excitation light.

3 EXPERIMENTS

3.1 Sensor Property

3.1.1 Wavelength Dependency

We investigated output current of the devised sensor

for several wavelengths. The sensor was exposed to

a range of wavelength lights from approx. 430 nm to

610 nm using programmable light source equipment

(OSVIS-500, Tokyo Instruments, Inc.) in a dark

room. Each power of the incident light was prelimi-

narily measured using an external power meter

BIODEVICES 2012 - International Conference on Biomedical Electronics and Devices

186

(8230E, ADC Corporation) to keep fixed light pow-

er of 100 μW/cm

for the reference.

3.1.2 Detection Limit

On the assumption of FITC use, the sensor was

irradiated by commercial light emitting diode (LED)

(L-7113VGC-H, Kingbright Elec. Co., Ltd.) that had

the dominate wavelength of 525 nm and the output

current was measured at 27 °C. Incident light power

was changed from 0.1 to 100 µW/cm

confirmed by

an external power meter (8230E, ADC Corporation).

3.1.3 Temperature Dependency

We investigated the influence of dark current which

comes from thermal excitation. The surrounding

temperature of the sensor was set 21, 27, and 35 °C

independently. Then, output currents were measured.

3.2 Fluorescence Detection

3.2.1 Measurements of FITC and Texas Red

To confirm the sensor’s performance of the quantita-

tive measurements for fluorescent dyes, fluorescein

isothiocyanate (FITC) and sulforhodamine 101 acid

chloride (Texas Red) which are typically used bio-

analysis including SNP genotyping, we measured

the excited fluorescence intensity 525 nm of FITC

(Wako Pure Chemical Industries, Ltd.) or 615 nm of

Texas Red (Life Technologies Corporation) 200 μl

solution of 0, 1, 10 μM individually. Each fluores-

cent solution was prepared using diluted ethanol

(50v/v%). And the each sample solution was set in a

cylindrical plastic chamber of a 7 mm inner diameter

and a 10 mm height as shown in Fig. 5. The excita-

tion lights, 490 nm and 590 nm, for FITC and Texas

Red were independently applied by light source

equipment (LAX-C100, Asahi Spectra Co., Ltd.).

The absorption depth of the sensor was varied by

changing of photo-gate voltage. The intensity of the

fluorescence light and excitation light can be calcu-

lated by using equations (1) where λ

= 490 nm, α

20.0×10

-1

, w

= 0.761 µm, and λ

= 525 nm, α

12.7×10

-1

, w

= 1.135 µm were substituted in the

case of FITC use. In the same way, λ

= 590 nm, α

6.39×10

-1

, w

= 0.803 µm, and λ

= 615 nm, α

4.82×10

-1

, w

= 1.135 µm were substituted in the

case of Texas Red use. The results of detected the

light intensities were normalized assuming the inten-

sity was 1.0 that was a value in the case of measur-

ing 10 μM fluorescent solutions. Furthermore, we

evaluated the possibilities of cross talks of two exci-

tation lights, 590 nm and 490 nm, to FITC and Tex-

as Red. Each fluorescent solution was additionally

exposed by ineffective light for excitation (i.e., 590

nm light for FITC and 490 nm for Texas Red).

Figure 5: Schematic view of the measurement setup.

3.2.2 Detection of FITC and Texas Red Mix

As a model sample of a heterozygote in SNP varia-

tions, we additionally measured the 200 μl mixture

solution of equal parts of FITC and Texas Red (net

0.5 μM for each). The results of detected light inten-

sities were also normalized as same as above section

experiment.

4 RESULTS AND DISCUSSION

Figure 6 shows that the correlation between wave-

length and output current of the sensor. The result

means that the sensor has high sensitivity near the

center of 550 nm at least in this wavelength region.

And also, the sensor has almost the same sensitivity

level for emission wavelengths of FITC (525 nm)

and Texas Red (615 nm).

We think that the similarity of the sensor’s out-

put current level of FITC and Texas Red will enable

us to compare them without any other complicated

calibrations in this study.

Figure 6: Wavelength dependency of the sensor's output.

Figure 7 shows the detection current versus the

incident light intensity. Each data point was obtained

by 10 times experiments and the each error bar was

PROPOSAL FOR A FILTERLESS FLUORESCENCE SENSOR FOR SNP GENOTYPING

187

standard deviation. Fig. 7(a) and (b) show the inci-

dent light intensity in the range of 2-100 µW/cm

and 0.1-2 µW/cm

, respectively. The detection cur-

rent is proportional to the incident light power in the

range at least above about 0.1 µW/cm

. The detec-

tion current was nearly equal to the dark current less

than 0.1 µW/cm

. This result implies that the sensor

can almost linearly detect incident light in the range

of 0.1-100 µW/cm

(a)

(b)

Figure 7: Detection current versus the incident light inten-

sity in the range of 2-100 µW/cm

(a) and 0.1-2 µW/cm

(b).

Figure 8 shows the dark current level depending

on the temperature. The result shows that dark cur-

rent increases with increasing of surrounding tem-

perature. According to the dark current level, we

found that our sensor is able to detect the fluores-

cence intensity quantitatively in the range of more

than about 0.1 µW/cm

. And it is presumed that the

dark current level is a few pA at room temperature.

The results of individual measurements of FITC

and Texas Red solution are shown in Fig. 9. Each of

them can be quantitatively detected in these rough

differences in concentration. It implies that the sen-

sor will be able to use for SNP genotyping because a

typical concentration of polymerase chain reaction

product is about 10 µM.

Figure 8: Temperature dependence of the dark currents of

the devised sensor.

(a) FITC solution.

(b) Texas Red solution.

Figure 9: Concentration dependency of detected fluores-

cent intensities.

(a) FITC solution.

(b) Texas Red solution.

Figure 10: Differences of fluorescence intensity by the two

different lights excitation.

0 20406080100

Output current [nA]

Incident light intensity [µW/cm

]

0.03

0.06

0.09

00.511.52

Output current [nA]

Incident light intensitey [µW/cm

]

0.002

0.004

0.006

0.008

0.01

21 27 35

Dark current [nA]

Temperature [°C]

0.0

0.5

1.0

0110

Light intensity

(525 nm) [arb. unit]

Concentration of FITC [μM]

0.0

0.5

1.0

0110

Light intensity

(615 nm) [arb. unit]

Concentration of Texas Red [μM]

0.0

0.5

1.0

0110

Light intensity

(525 nm) [arb. unit]

Concentration of FITC [μM]

Excitation = 490 nm

Excitation = 590 nm

0.0

0.5

1.0

0110

Light intensity

(615 nm) [arb. unit]

Concentration of Texas Red [μM]

Excitation = 490 nm

Excitation = 590 nm

BIODEVICES 2012 - International Conference on Biomedical Electronics and Devices

188

Figure 11: Distributions of fluorescence intensities for

FITC solutions, Texas Red solutions and their mixtures.

Ideally, those mixtures should be measured to the

center of both axes. However, they slightly leaned to

FITC side. We suppose that this result comes from

FITC appears large seemingly due to fluorescence of

Texas Red which was excited by 490 nm light is

mixed. According to these results, we think that the

devised sensor could potentially be applied to real

SNP genotyping for individual determination of

homozygote or heterozygote.

5 CONCLUSIONS

We showed that our devised sensor could detect

quantitatively fluorescent dyes; FITC and Texas Red

without the filters. And, we found that the sensor

successfully distinguished FITC and Texas Red

solution and their mixture. We believe that the sen-

sor enable us to SNP genotyping in compact system.

ACKNOWLEDGEMENTS

This work was partially supported by the Global

COE Program titled Frontiers of Intelligent Sensing,

the projects of the Hamamatsu Optoelectronics

Knowledge Cluster Initiative, a grant-in-aid for

scientific research (A) (No. 21246051) from the

Ministry of Education, Culture, Sports, Science and

Technology (MEXT) of Japan, and a grant-in-aid for

JSPS Fellows from Japan Society for the Promotion

of Science (JSPS).

REFERENCES

Y. Maruyama et al., IEEE Trans. Electron Devices, 53(3)

(2006) 553–558.

Y. Maruyama et al., Sens. Actuators A 128 (2006) 66-70.

H. Nakazawa et al., Proc. Transducers 2011 (2011) 100-

103.

J.P. Mckelvey, Solid State and Semiconductor Physics,

Harper and Row, New York, 1966, p. 463.

0.3

0.6

0.9

1.2

00.30.60.91.2

Texas Red fluorescence light

intensity rate [arb. unit]

FITC fluorescence light

intensity rate [arb. unit]

▲ Only Texas Red

■ Mixture

◆ Only FITC

PROPOSAL FOR A FILTERLESS FLUORESCENCE SENSOR FOR SNP GENOTYPING

189