Microfluidic device for rapid, automated DNA extraction for PCR based diagnosis

Adil Rehman

School of Engineering Sciences and Technology (SEST), Jamia Hamdard, New Delhi, Delhi 110062

Prof. Sai Siva Gorthi

Department of Instrumentation and Applied Physics, Indian Institute of Science (IISc), Bengaluru, Karnataka 560012

Abstract

Plant diseases are a serious and continuing threat to food security as well as the national economy. While there are several techniques to detect such disease, out of that, molecular biology-based ones particularly Polymerase Chain Reaction (PCR) are the most accurate and sensitive. These methods differ from others in their selective targeting and amplification of DNA/RNA fragments belongs to the pathogen within a short time. However, PCR still isn't used very widely, partly because of the high operator skill needed and partly because of the costly & bulky instruments required. This research is aimed at resolving a part of this problem. It specifically targets the steps related to sample preparation, which is one of the three sub-steps of the reaction (the other two being sample processing and then detection). This step is the most challenging since it requires precise volumetric handling and contamination-free processing across multiple sub-steps. One of the most efficient and easy-to-use techniques for sample preparation (for PCR) is magnetic beads-based DNA extraction (MBbDE). This research aims to implement the MBbDE in an automated manner via a microfluidic device and digitally controlled actuation. The microfluidic device ensures precise volumetric processing while the low-cost digital control provides automation. The microfluidic device has three microchannels that are interconnected with three vials that contain samples and reagents. Through a digitally controlled actuation prototype, we accomplished valving of fluid, suction & traction of fluid via inlets & outlet of a microfluidic device. This ultimately results in the mechanization of processes involve in MBbDE such as lysing, binding, washing, and elution. It is efficient, user-friendly, reduces contamination along with consumption of samples and reagents. Moreover, it also provides accurate DNA samples as end-results, which are further used for the identification of pathogens.

Keywords: Molecular biology, PCR, Microfluidics, PDMS, MBbDE, Valving

Abbreviations

INTRODUCTION

Background

Agricultural productivity is something on which the economy of India highly depends. This is one of the main reasons that disease detection in plants plays an important role in the agriculture field, as having the disease in plants is quite natural. If proper care is not taken in this area then it causes serious effects on plants and due to which respective product quality, quantity, or productivity is affected ^[1] .​​ For instance, a disease named little leaf disease is a hazardous disease found in pine trees in the United States. Detection of plant disease through some automatic technique is beneficial as it reduces a large work of monitoring in big farms of crops, and at a very early stage itself it detects the symptoms of diseases i.e. when they appear on plant leaves​ ​^[2]​ . Visually identifying plant diseases is inefficient, difficult, time-consuming, requires expertise in plant diseases and continuous monitoring which might be expensive in large farms​ ​^[3]​ . Therefore, a fast, automatic, and accurate method to detect plant disease is of great importance. The disease which we focus on in this research is Bacterial Wilt.

Bacterial Wilt

Bacterial wilt is a tomato disease that is caused by the pathogen bacterium R. solanacearum. It is very common in moist sandy soils and humid environments, such as the conditions of the coastal south of the US ^[4]. The bacteria reside in the soil and work their way very quickly through the roots and up the stem of the plant. Bacterial wilt is often the result of a plant being injured, cut, or weakened by insects or wear and tear from handling ^[5] . When infected, the bacteria builds up and clogs up the stems of the plant, keeping water and nutrients from being able to reach the leaves, effectively killing the plant ^[6] .

Causes and Symptoms of Bacterial Wilt

The youngest leaves show signs of infection first and begin to wilt during the hottest part of the day. This often goes unnoticed as the leaves stay green during the infection ^[7]. Eventually, the wilting will become obvious, but once we notice it happening, it is likely that the entire plant has begun to wilt and will soon die​ ​ ^[8]. Bacterial wilt tends to occur when the weather is extremely hot and there is a high level of humidity from recent rainfall, leaving the soil wet. Bacterial wilt is also common in areas with a high soil pH ^[9]. Bacterial wilt can be diagnosed by cutting the stem at the base of the plant and searching for discolored tissue. Suspend suspect stems in a glass of water to test for bacterial wilt. Infected stems will ooze a slimy, white substance into the water within minutes of being submerged ​^[10]. The symptoms of Bacterial Wilt are shown in Fig. 1.

Fig 1 A symptom of bacterial wilt of tomato caused by R. solanacearum showing wilting of foliage and stunting of the plant.

Treatment and Control of Bacterial Wilt

There are no known effective chemical controls for bacterial wilt. As the plants die, the bacterial pathogen is released into the soil, so the most important thing you can do to prevent the spread of bacterial wilt is to remove diseased plants as soon as you notice wilting ^[11]. Do not compost these plants, but discard them immediately and clean and sanitize any tools used in the removal and disposal process ^[12].

So, since there is no way to cure infected plants, the best way to control bacterial wilt is to practice prevention methods. Use good cultural controls to keep bacterial wilt out of your garden and off of your tomatoes​​. Few existing ways are using which we can prevent bacterial wilt issues namely-rotating crops regularly, install raised beds, space plants out evenly to improve air circulation, testing soil and amend to a pH of 6.2 to 6.5 for tomatoes and most garden vegetables, wash hands and gardening tools after handling infected plants, if problems persist with a soil-borne disease we have to try shifting to container gardening using a sterile commercial potting mix ​​​​^[13]​​​.

However, the above method is not too effective as well as accurate for the prevention of plants from the pathogen R. solanacearum that causes Bacterial Wilt. Therefore, in this research, we used molecular biology-based techniques (particularly PCR). It is different from others in terms of targeting and amplification of DNA/RNA fragments that belong to the pathogen within a short time. However, PCR still isn't used very widely, partly because of the high operator skill needed and partly because of the costly and bulky instruments required. This research is aimed at resolving a part of this problem. It specifically targets the steps related to sample preparation, which is one of the three sub-steps of the reaction (the other two being sample processing and then detection). Moreover, it will provide accurate DNA samples as end-results, which are used in further steps for the identification of pathogens. This problem of Bacterial Wilt should be solved at the early stage because if not then it infected the nearby soil area as well as other plants that are there in their neighborhood. This contamination of soil also affects the yield rate of vegetables (specifically tomatoes). In addition to this, it is having the characteristic of lethality, persistence, wide host range, high surviving ability, and broad geographic distribution.

Statement of the Problems

In the process of detecting the pathogens for Bacterial Wilt, we used the PCR technique. This PCR reaction involves three steps namely-sample preparation, sample processing, and detection. Out of these three, the most intricate one is sample preparation (DNA samples) because it requires precise volumetric handling and contamination-free processing across multiple sub-steps. The sample preparation step acts as a base for the rest of the reaction. Therefore, if any kind of error is involved in this step, then it ultimately leads to false results in terms of detection of the pathogens which is again destructive for the plants as well as soil. The most efficient and easy-to-use technique for sample preparation (for PCR) is MBbDE. However, it is still a conventional process that requires high skill operator, close monitoring of each process is also needed.

Objectives of the Research

The main objective of this research is to implement the MBbDE in an automated manner via a microfluidic device and digitally controlled actuation for the rapid extraction of DNA samples, which is further used in the detection of pathogens.

Overall objective

The overall objective of this research is divided into several sections, which are:

• Design & fabricate a Microfluidic device using PDMS and glass substrate with three inlets and one outlet with specified dimensions for precise volumetric handling.
• Fabricate a low-cost digitally controlled actuation prototype using a linear stepper motor, servo motor attached with spring-loaded mechanism, and microcontroller unit that serves the automation part.
• Interfacing of Microfluidic device and digitally controlled actuation prototype to automate the whole process involved in MBbDE which is used for the extraction of DNA samples.
• Qualitative and quantitative analysis of DNA samples w.r.t on-chip DNA extraction method.

Scope

For a higher agricultural productivity rate on which the economy of India mostly depends, disease detection in plants plays an important role. Hence, employing this research we can identify the pathogens in plants by extracting DNA samples through a microfluidic device interfaced with a digital control actuation prototype. This results in early-stage identification of disease, higher agricultural yield rate, contamination of soil & other plants should also be avoided.

LITERATURE REVIEW

Information

Several works had been done in the field of molecular biology specifically w.r.t automation of DNA extraction process which is further used in the detection of disease in plants and vegetables, e.g., Kim et al. proposed an automation DNA extraction system that is solely based on microfluidics & magnetic bead. This whole system comprises of microcontroller, four linear stepper motor, cartridge with chambers, stepper actuator, and PC-based host application. This host application is responsible for controlling the stepper motor to perform the various process of magnetic bead automatically. However, the major drawback of this system is its cost, compactness, dependency on another system to perform its operation ^[14]. In another study, Gan et al. incorporated a plastic microfluidic device that is linked with a filter disc to extract the DNA samples automatically. This whole system is also connected with a syringe pump, and MVP valve to achieve automation. Although this whole system is low-cost and rapid, there are some limitations, such as lack of rigidness, fragile due to the filter disc is made up of filter paper that can be easily deteriorated ^[15]. Montpetit et al. also carried out work in DNA extraction automation by fabricating a simple automated instrument,i.e. Qiagen BioRobot EZ1, which is a small, rapid, and reliable automated DNA extraction instrument capable of extracting DNA from up to six samples in as few as 20 min using magnetic bead technology. Still, there is a major downside that is attached with this instrument: namely, the need for expert users to operate the instrument, expensive and insubstantial ^[16]. Mehle et al. utilized a different approach for the automation of DNA extraction by combining a simple and quick homogenization step of crude extracts with DNA extraction based upon the binding of DNA to magnetic beads. Nevertheless, there is some significant demerit also like lack of compactness, high consumption of chemicals & reagents requires a huge amount of time to complete the process ^[17].

Summary

To counter the limitations of the earlier research, we fabricated and designed a microfluidic device that is interfaced with digital controlled actuation comprised of a linear stepper motor, microcontroller, and servo motor linked with a spring-loaded mechanism to automate the whole process involved in MBbDE that is used in the extraction of DNA samples. This whole system does not require any technical expertise from the user to perform its operation. It is modular w.r.t design, user friendly, low-cost, compact, and durable . It also avoids the wastage of chemicals and reagents for its functionality. It is independent of any third-party instrument.

METHODOLOGY

The methodology section of this research is divided into two sections:

• How DNA is collected?
• How DNA is analyzed?

Collection of DNA samples

For the collection of DNA samples there are two devices that we fabricated and designed:

• Microfluidic Device
• Digitally Controlled Actuation Prototype

Microfluidic Device

The designing of a Microfluidic device is done through software named as CleWin 4.0 where we made a design with specified dimensions (25 mm x 75 mm) and geometries. Then we made a photomask of it. The design as well as the photomask is shown in Fig. 2.

1

2

3

4

5

6

7

8

a) Photomask design in CleWin 4.0 

• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8

b) Photomask used in the Fabrication process of Microfluidic device

Fig 2 Illustration of design for Microfluidic device

After the designing part, we had done the fabrication of the Microfluidic device using the above photomask. There are several steps involved in fabrication, which is majorly divided into two parts:

• Construction of Microfluidic Master using Photolithography: In this, we first take a circular glass substrate and clean its surface with isopropanol & acetone to remove all the unwanted foreign particles that are there on its surface. Then we leave it for some time with the aim that it becomes dry. This is shown in Fig. 3a. In the meantime, we prepare a photo-imageable etching resist ink (with its hardliner and main agent) in the ratio of 1:3. It is a type of negative photoresist. It is clearly shown in Fig. 3b. Then we take the glass substrate and put it inside of the spin coating machine on a provided circular base mount precisely. Poured the photo-imageable ink on the top surface of the glass substrate and turn on the vacuum pump of the spin coating machine such that it holds the glass substrate while it is in rotation. Then set the RPM of the spin coating machine at 3500 RPM (It needed to be set because this decides the thickness of coated ink on glass substrate) with a timer of 90 seconds. It is shown in Fig. 3c. After the spinning process is done whole photo-imageable etching ink is distributed evenly on the surface of the glass substrate. It is represented in Fig. 3d. Now, we take coated glass substrate and put it inside of hot air oven at a temperature of 65^o centigrade for approximately about 20 minutes to solidify the photo-imageable ink that is already coated on a glass substrate. It is depicted in Fig. 3e. After this, we take out coated glass substrate put it under some black container (Due to photo-imageable ink is sensitive to the part of the visible light spectrum) for 5 minutes to avoid any error in printing the pattern of the photomask on it because of some non-solidified portion on the coated surface. Then take the photomask, put it precisely on the surface of a coated glass substrate and press it moderately to achieve contact lithography. Then, expose it in a UV cabinet by turning on the long UV(approx. 365 nm) for about 15 minutes with the intention that the exposed portion to light becomes insoluble to the photoresist developer while the unexposed portion of the photo-imageable coated glass substrate is dissolved by the photoresist developer. It imprints the pattern of photomask on the surface of coated glass substrate This is portrayed in Fig. 3f. Then we take out the coated substrate that is in contact with the photomask on the top of it and peel off the photomask from the substrate. Then, we put the coated substrate in the washtub containing photoresist developer solution of sodium carbonate & water with the ratio of 100:1. Stir the tub gently. After some time the pattern of the photomask is visible on the glass substrate. Rinse it with water then put the substrate on blotting paper for some time so that water gets absorbed. This patterned coated glass substrate is called master which we further used in the fabrication of a Microfluidic device. This is acutely represented in Fig. 3g. Finally, we check for any damage in terms of the pattern imprinted on the master with the help of a microscope. It is shown in Fig. 3h.

a) Cleaned glass substrate

b) The photo-imageable ink

c) Before coating photo-imageable ink

d) After coating photo-imageable ink 

e) Coated glass substrate inside of an oven

f) UV exposure after placing photomask over coated glass

g) Microfluidic master with imprinted pattern of photomask 

h) Verification of any damage in the imprinted pattern

Fig 3 Demonstration of the creation of microfluidic master using photolithography 

• Fabrication of Microfluidic device utilizing soft-lithography:  After the construction of the master, we used that same master for the fabrication of the microfluidic device through soft-lithography. Here we first put it in a plastic petri dish with some 4cm depth. Use waste PDMS blocks (solidify one) and put that one in the dish containing master in such a way that only target pattern rectangular boundary is not covered. This is important because by using this technique we can further minimize the usage of elastomer base and elastomer curing agent which we ultimately used for the preparation of the PDMS device. It is clearly shown in Fig. 4a. After this, we prepared the mixture of elastomer base & elastomer curing agent in the ratio of 10:1 by mixing it rigorously for about 10 minutes. It is portrayed in Fig. 4b. Then we desiccated the bubbles which are formed due to the mixing process inside of a desiccator for approx. 10 minutes. This step is done 3-4 times to get rid out of maximum bubbles (if not eliminated causing major problems in a microfluidic device in terms of patterns). This is shown in Fig. 4c. Now, we poured the mixture onto a dish containing master with waste PDMS blocks. Filled it precisely according to our thickness requirement. It is clearly shown in Fig. 4d. We take the dish containing the mixture and cured it for exactly 2 hours at 65^o centigrade using a hot air oven with the aim that it becomes solid along with the pattern of the master is embossed on the bottom side of the PDMS device. After 2 hours we take out the dish from the oven and leave it for 5-10 minutes to cool the dish. It is shown in Fig. 4e. Now, we peeled off only the target rectangular pattern PDMS device and observed that on the bottom side of the PDMS device we had embossed microstructures patterns. Then, we punched the 1mm holes using biopsy punches with a plunger system at the target location. This should be done at the bottom side where the pattern is embossed. After that, we paste the scotch tape at the bottom side along with it we also take a glass slide of dimension (25 mm x 75 mm) where we put the scotch tape on the top surface of it as well. From both rectangular pattern PDMS device and glass slide, we removed the scotch tape as well as cleaned them by rinsed them with isopropanol & acetone to remove the unwanted particles. Then dry them out for 1 minute by putting them on the non-linkable tissue. It is shown in Fig. 4f. Then, both the PDMS device and glass slide are put into the chamber of the Harrick Plasma Cleaner for plasma bonding between the glass slide & PDMS device. It is depicted in Fig. 4g. Then, we turn on the plasma cleaner, oxygen pump, and vacuum gauge. Close the lid of the chamber by closing its knob in the downward direction and checked whether it is completely closed by observing either vacuum is created or not. It is shown in Fig. 4h. Now, we observed the reading of the vacuum gauge and when it comes below 100 then first turn the knob in the left direction slowly, open the oxygen supply valve until the reading of the vacuum gauge comes in between 700-800(close the knob by turning it in the downward direction). Again, we monitored the reading of the vacuum gauge when it comes below 100 then we turn on the power supply for the radio frequency ejector and sets it to knob at high. Then observed the chamber when it glows to pink, plasma bonding process happened. This is clearly shown in Fig. 4i. We waited for 3 minutes then turn off the pump, close the power supply of the radio frequency ejector as well as sets its knob to low. Then observed the reading of the vacuum gauge, when it comes in between 600-700 turned the knob of the plasma cleaner in a right direction slowly to remove the vacuum inside of plasma cleaner chamber. Immediately take the glass slide from the plasma cleaner chamber and put the PDMS device on the top of it by closely checked whether the pattern of the embossed microstructures is in contact with the surface of the glass slide or not. Pressed it gently so that no air gap remains there. Then immediately put the combined PDMS device and glass slide on the hot plate for about 20 minutes at a temperature of 80^o centigrade. It is shown in Fig. 4j. After that, we finally fabricated our desired Microfluidic device. It is depicted in Fig. 4k.

a) Petri Dish containing master and waste PDMS blocks

b) A mixture of elastomer base and curing agent for the preparation of PDMS device

c) Desiccation of bubbles that are formed in mixture inside a desiccator

d) Poured the mixture on a petri dish containing master and waste PDMS blocks

e) Cured PDMS device

f) Punched PDMS device and a glass slide

g) PDMS devise & glass slide inside the chamber of Harrick Plasma Cleaner

h) Closed the chamber of Plasma cleaner to check whether a vacuum is created or not

i) Plasma bonding process going on inside the chamber of the plasma cleaner

j) Bonded PDMS device and glass slide on the top of a hot plate

k) The fabricated  microfluidic device

Fig 4 Fabrication of microfluidic device via soft-lithography

Digitally Controlled Actuation Prototype

The fabrication of a digitally controlled actuation prototype is made up of a linear stepper motor (which we used as a syringe pump by connecting it with a lead screw and 3D printed casing where we can fixed syringe), servo motor attached with spring-loaded mechanism, microcontroller. For the fabrication of the syringe pump we first 3D printed our design then connected all the required components. After this, we calibrated our syringe pump by initial testing with a known amount of samples. It is shown in Fig. 5.

1

2

3

4

• 1
• 2
• 3
• 4

Fig 5 Illustration of 3D printed syringe pump

Then we set up a valving system that is needed for blocking the channels of our Microfluidic device. It is done by utilizing a servo motor attached with a spring-loaded mechanism mounted on an optical workbench. After this, we combined our 3D printed syringe pump and valving system. Then we programmed this whole system by connecting it with a microcontroller. It is depicted in Fig. 6. This whole system is what we called our digitally controlled actuation prototype that serves the automation process.

1

2

3

• 1
• 2
• 3

Fig 6 Digitally controlled actuation prototype

After this, we integrated our microfluidic device (which is stuck with a magnet on the top of it) and digitally controlled actuation into one system that is used for the extraction of DNA. Then, we connected the samples i.e. cocktail sol., wash buffer, and elution buffer which is stored inside of separate vials connected to the microfluidic device using polyethylene tubings with inlet-2, inlet-3, inlet-1 of the microfluidic device respectively, and outlet of the microfluidic device is connected with a syringe pump attached with 2.5 ml syringe through a separate tubing. Afterward, we turn on the system for the extraction of DNA. It is shown in Fig. 7.

1

2

3

4

5

6

7

8

• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8

Fig 7 Display of the whole system which is used for the extraction of DNA

Algorithm

For a better understanding of the DNA extraction sequence, we described an algorithm below:

• When we turn on the system then firstly our valving system is at its rest position & all the channels of a microfluidic device are open. However, after 1 second all the channels are closed by turning the servo motor into the position where it exerts a full force on the spring-loaded mechanism.
• Next, at the serial monitor of our software IDE, we have to enter the value of certain parameters such as the diameter of the piston inside of the syringe, pitch of the lead screw, the total volume of the syringe, volume of the elution buffer, and the time of the process.
• The above value is used for the calculation of flow rate and steps required to displace the total volume of fluid through a cross-sectional area of a syringe, steps per second, and steps required to push a volume of elution buffer.
• After this, a maximum speed of rotation of the syringe pump is set by steps per second.
• Then, our main system performs its operation by first creating a suction in the syringe as well as a microfluidic device connected with vials.
• After this Channel-2 is open while the other two channels are blocked using the valving mechanism. Cocktail solution (650 µl) got completely sucked out by syringe pump for the duration of 1 hour via microfluidic device & its outlet. During this process the DNA particles that are linked with magnetic beads inside of cocktail sol. got attached to the top surface of the cavity inside the microfluidic device because of the magnetic field created by the magnet.
• Then Channel-3 is open while the other two channels are blocked using the valving mechanism. Wash buffer solution (100 µl) got completely sucked out by syringe pump for 25 minutes via microfluidic device & its outlet. This step washed out all the supernatant particles linked with DNA except magnetic beads. Afterward, we drew out some amount of air from the same channel for about 5 minutes.
• Next Channel-1 is open while the other two channels are blocked using the valving mechanism. Elution buffer solution (30 µl) got completely sucked out by syringe pump for 15 minutes via microfluidic device & its outlet. This step split up the magnetic bead and DNA particles. Here magnetic bead settled on the top surface of the cavity & also loses its characteristic to make a bond with DNA particles. On the other hand, DNA particles go into the syringe, but due to the air gap, they do not mix with other solutions inside of the syringe.
• Now, all the channels are closed for 5 seconds. Afterward, only Channel-3 is open.
• We pushed the fluid containing DNA inside of syringe with a flow rate of 1 ml/hr using a syringe pump via a microfluidic device. However, the syringe pump performed this operation till it reached the required steps that we calculated earlier for the displacement of total elution buffer volume.
• Finally, in vial-3 which is connected with Channel-3, we obtained our solution that contained the DNA particles.

Analysis of DNA samples

After the collection of the DNA samples into the vial in the form solution, we calculated the amount of concentration by measuring its intensity of light detected. It is calculated using spectrophotometer. We first run on this instrument and the dsDNA (double strand DNA) option. Then clean the target region (control space inside of spectrophotometer) to remove the impurity where we further drop the sample. Run it once to complete the calibration. Afterwards, we take 2 μl of milli-Q water into the pipette and drop it into the target region. Close the lid by checking it from side view to see whether a bubble is formed or not. Select blank run option. When it gets finished only then we will be able to use measure option. This is the blank run without any sample. If we correctly do this then there should be no readings shown by spectrophotometer.

Then, we take 2 μl of DNA sample solution into the pipette & drop it onto the target region of the spectrometer. Select measure option, after the process is completed we will be able the amount of DNA concentration in ng/μl, along with this we also have the readings for the quantity of nucleic acid (A260), purity of DNA and RNA (260/280), and secondary measure for purity of RNA or DNA (260/230). The ratio of 260/230 is the absorbance wavelengths (in nm) used to assess the purity of DNA and RNA. A ratio of 1.7-2.0 is considered pure for DNA and a ratio of ∼ 2.0 is considered pure for RNA. A lower absorbance ratio may indicate the presence of protein, phenol or other contaminants that have an absorbance close to 280 nm. On the other hand, the ratio of 260/230 can be used as a secondary measure of DNA or RNA purity. In this case, a ratio between 2.0-2.2 is considered pure. If the ratio is lower than this expected range, it may indicate contaminants in the sample that absorb at 230 nm. When the instruments done its operation on the sample, we get the quantitative readings of DNA in the form of concentration. In addition to this, a graph is also generated between wavelengths vs. absorbance. This graph is significant because its describe that at what wavelength DNA particles has the highest absorbance property which also explained the maximum concentration of DNA should also be there on this particular wavelength. The other parameters describe purity of DNA as well as presence of contamination in the DNA samples. The spectrophotometer while performing its operation is shown in Fig. 8.

1

2

3

4

a) Overview of parameters

• 1
• 2
• 3
• 4

b) Overview of the whole instrument

Fig 8 Demonstration of a Spectrophotometer

RESULTS

Validation with Ink

We first validate the working of the whole system w.r.t. dyes i.e. methylene blue & methylene red and milli-Q water. Here we take the methylene blue, methylene red, and milli-Q water of known amount of volume into the vail-1, vial-2, vial-3 respectively. Then we perform the sequence of suction, blocking & opening of a channel by the algorithm that we discussed earlier. We observed that all the steps run smoothly and we got the dyes into the syringe accurately as well as by the end of the operation the required volume of fluid is also completely pushed at our predefined push rate via syringe pump connected with a 2.5 ml syringe through the microfluidic device. It is stored in vial-3. The demonstration of this validation is shown in Fig. 9.

Fig 9 Validation with milli-Q water and Dyes

Afterward, we performed the extraction of DNA with the same whole system except microfluidic device & polyethylene tubing to avoid any contaminations while collecting the DNA samples. We had taken 650 µl of the cocktail solution including tomato sample in vial-2 (connected with inlet-2), 100 µl of wash buffer solution in vial-1 (connected with inlet-2), and 30 µl of elution buffer in vial-3 (connected with inlet-3) respectively. Run the whole system for about 1 hour 45 minutes by constantly monitoring each step of suction, blocking as well as opening of a channel through syringe pump & valving mechanism. We observed that all the steps that we discussed earlier in form of the algorithm run accurately without any disturbances. In the end, we perceived that collected the solution that contained the DNA particles into vial-3. The exposition of this extraction of DNA is shown in Fig. 10.

Fig 10 Extraction of DNA samples in the form of solution

After the collection of DNA samples which is present in the solution stored inside of vial-3, we analyzed it using spectrophotometer to detect the concentration & purity of DNA. The result of this experimental sequence is shown in Fig. 11.

1

2

• 1
• 2

Fig 11 Quantitative outcomes of the DNA samples extracted from Tomato

As we clearly see, the concentration of extracted DNA from tomato sample is 1.801 ng/μl, which is in the acceptable range specifically for DNA amplification & replication through PCR (we can use this technique to create millions of copies). These additional copies of DNA can be used for the further detection of pathogens by amplifying the target DNA region. The extracted DNA has maximum absorbance at a wavelength of 250-255 nm. In addition to this, there is some amount of protein contamination present in the tomato DNA sample.

We also performed the DNA extraction operation with same whole system except microfluidic device and polyethylene tubing. This time we had taken 650 μl of the cocktail solution including chilly sample in vial-2 (connected with inlet-2), 100 μl of wash buffer solution in vial-1 (connected with inlet-2), and 30 μl of elution buffer in vial-3 (connected with inlet-3) respectively. Run the whole system for about 1 hour 45 minutes by constantly monitoring each step of suction, blocking as well as opening of a channel through syringe pump & valving mechanism. We observed that all the steps that we discuss9ed earlier in form of the algorithm run accurately without any disturbances. In the end, we noticed that the collected solution contained the DNA particles into vial-3. The analysis of collected DNA samples, which is present in the solution, stored inside of vial-3 is executed using spectrophotometer to detect the concentration & purity of DNA. The outcomes of this sequence is shown in Fig. 12. If we observed the result closely then we see that there are four DNA samples concentration, but all of them came from the from the Chilly sample by performing the elution buffer step 4 times with separate 30 μl of Elution Buffer solution. For the third (since we performed elution buffer steps several times) chilly sample we achieved the highest amount of concentration of DNA i.e. 3.564 ng/μl. This is in acceptable range for replication & amplification of the DNA, which ultimately helps in the identification of pathogens. The extracted DNA has maximum absorbance at a wavelength of 220-240 nm. Moreover, there is some amount of protein contamination present in the chilly DNA sample. The whole result of this operation is represented in Fig. 12.

Fig 12 Results of the DNA samples extracted from Chilly

Gel Electrophoresis of R. solanacearum culture samples extracted by microfluidic device

• The bacterium R. solanacearum infect tomato and chilly plants were used for extracting DNA by microfluidic device.
• The extracted DNA was amplifying used a thermal cycler. The temperature profile are as follows- Initial denaturation- 96⁰C for 2min, denaturation- 94 ⁰C for 20s, annealing- 68 ⁰C for 25s, extension-72 ⁰C for 30s. The reaction runs for 35 cycles with total reaction volume of 20 μl.
• After amplifying the DNA on thermal cycler, the PCR products were tested on 2% Agarose gel. The bands were analyzed on gel documentation system
• The desired PCR product should be 292bp length as per the literature. The result of gel electrophoresis is shown in Fig. 13.

Fig 13 Results of the Gel Electrophoresis for tomato and chilly sample

CONCLUSION

In a nutshell, we achieved our objective i.e.to create a low-cost prototype that is fabricated by the unification of microfluidic device & digitally controlled actuation for rapid, automated DNA extraction. In this, the user does not need to monitor the whole process. He/she only needs to put the required samples in the vials and turn on the prototype. In the end, he/she got the DNA particles, which are further used for the detection of pathogens. It ultimately reduces the dependency of the whole system on the user which makes it user-friendly as well as the user does not need to have any expertise while dealing with this system. It also decrements the chances of any error, which we commonly noticed in the off-chip DNA extraction method specifically due to the carelessness of the user who performed the whole operation using the conventional MBbDE method. In addition to this, it solves the most complex problem that researchers find in the sample preparation step that is precise volumetric handling and cross-contamination of the samples. It also utilizes the samples and reagents efficiently. Furthermore, there are chances of improvement in this prototype like constructing a feedback system integrated with it to monitor the fluid position inside of microfluidic device by using image processing. The design of a microfluidic device will also become more efficient if we somehow either reduce the fluidic resistance or distribute it evenly. The efficiency of this prototype can also be increased if we will be able to make it completely airtight. A user interface will also be connected with it to create it more user-friendly when they enter the value of certain parameters, which is required to run the syringe pump. It will also have the possibility to produce a more compact design of the whole prototype using custom 3D-printed casing.

ACKNOWLEDGEMENTS

I feel thankful to get selected for the Indian Academy of Sciences’ Summer Research Fellowship Program (SRFP 2020) and pursue my research fellowship project at the Indian Institute of Science, Bengaluru. I am beholden to the Indian Academy of Sciences for giving me this golden opportunity and this had surely been a great learning experience for me to work in such a renowned institute.

I express my sincere profound gratitude to my guide, Prof. Sai Siva Gorthi, OMI labs, Department of Instrumentation and Applied Physics, IISc Bengaluru. It has been an honor to be his research intern for 8 weeks. I sincerely thank him for his excellent guidance, support, and motivation in every step throughout my research and during the whole period of my project.

I especially thank my mentor Mr. Prateek Katare and Ms. Roopa Ashwath, for their valuable support, patience, critical comments as well as a detailed review during the period of my project. They are the ones who always inspired me, were always willing to help, and give their best suggestions.

At last, I would like to express my deepest sense of gratitude towards my teacher Asst. Prof. Syed Sibtain Khalid, and my little sister Ms. Eram Rehman alongwith my family for their constant support & motivation without which I would not have been able to participate in this research fellowship program.