1) Here is the pyrosequencing concept taken from Manz, 2004, p. 136-139: A single stranded DNA template is immobilized onto a surface. (Personal comment: Mainz states here that the fragment is usually PCR amplified/purified while in the Margulies paper, the DNA is digested, bonded to the bead, and then amplified on the bead). Next, dATP, dGTP, dCTP, or dTTP is added to the reaction mixture one at a time (Klenow fragment, primer, ATP sulfyrase, luciferase, apyrase, APS, and luciferin are also in the reaction mixture). For example, dATP is added first. If A is complementary to the unknown strand, it will be added. If A is not complementary, apyrase will remove all of the dATPs by degrading it and the next base is attempted. If A is incorporated, inorganic phosphate (PPi) is released. The amount of inorganic phosphate is proportional to the number of sequential bases. For example, if there is 1 A added in a row, there will be 1 unit of PPi released. If there are 3 A's added in a row, there will be 3 units of PPi released.
APS (adenosine 5' phosphosulfate) then reacts with PPi via the enzyme ATP sulfurylase. The ATP sulfurylase reaction forms ATP and sulfate. This reaction is also reversible and may account for the name of the enzyme. I propose here that since there is no ATP in the reaction mixture, the reaction is favored to go forward towards the products.
The ATP then reacts with the substrate luciferin via the enzyme luciferase. This reaction produces AMP, phosphate, oxyluciferin, and light. The light is proportional to the amount of ATP (which correlates to the amount of inorganic phosphate and the amount of base added). By reacting one base at a time, the light (and its intensity) can be correlated back to the identity and quantity of the base.
2) Margulies et. al. stated that the time needed to fragment, prepare, and sequence the genome of 'mycoplasma genitalium' was 4 hours by one individual. I am not aware of how long it takes to do a Sanger sequencing experiment including the time it takes to prepare amplified colonies. Steinberg et. al. suggested that it was longer than the time necessary for the Margulies method.
3) The DNA is loaded into fiber-optic slides. The slides are then loaded into a flow chamber where sequencing reagents can flow above the pores. The chamber delivers reagents (more specifically bases) one at a time. The reagents are able to freely move into the chamber and react (or not react). Margulies reports the time frame of base reaction to be 10 seconds. ATP sulfurylase and luciferase action occurs in 0.02-1.5 seconds. When the correct base is added, the charge-coupled device (located in the closed part of the slide) detects emitted photons. After each reaction of base, apyrase is introduced to the slide to degrade any unused reagents. The slides are created with optic fiber etching.
I have a question that may seem completely dumb, but is there any inherent error to be considered since you are working with such small fiber-optic slides? It just seems like you would introduce significant error when working with this type of medium.
In response to Courtney's question, I believe the error can be minimal with the fiber optic slides. The slides are machined by fiber optic etching and the loading and analysis errors can be minimized by automating the process. If the slides were analyzed or loaded by hand, then this may introduce a significant possibility for error that must be accounted for.
I think the slide error could be correlated to the decrease in base error frequency. There is only a fixed volume inside the pore for which the bead containing fragmented DNA helicies can be placed. As DNA is extended onto the bead, there might be a limited ability for the DNA to extend OR for the reagents to enter the pore and react. Also if new reagents are not introduced into the system, the Klenow fragments could be degradaded, reagents could become limiting, and ions could be lost in the process.
The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of Minnesota.
Comments
Here are my questions and comments for Chad and the rest of the class.
1. Pyrosequencing - See Section 5.4 in Manz. We all need to know exactly how this method works (including enzymes such as sulfurylases).
2. Methodology - How long does it take to complete a pyrosequencing cycle?
3. Methodology -We need more detail on how the wells to conduct the pyrosequencing and to deliver the reagents were built.
4. Data Analysis - Figure 4, why does accuracy go down with base position?
Posted by: Edgar Arriaga | February 21, 2008 11:59 AM
1) Here is the pyrosequencing concept taken from Manz, 2004, p. 136-139: A single stranded DNA template is immobilized onto a surface. (Personal comment: Mainz states here that the fragment is usually PCR amplified/purified while in the Margulies paper, the DNA is digested, bonded to the bead, and then amplified on the bead). Next, dATP, dGTP, dCTP, or dTTP is added to the reaction mixture one at a time (Klenow fragment, primer, ATP sulfyrase, luciferase, apyrase, APS, and luciferin are also in the reaction mixture). For example, dATP is added first. If A is complementary to the unknown strand, it will be added. If A is not complementary, apyrase will remove all of the dATPs by degrading it and the next base is attempted. If A is incorporated, inorganic phosphate (PPi) is released. The amount of inorganic phosphate is proportional to the number of sequential bases. For example, if there is 1 A added in a row, there will be 1 unit of PPi released. If there are 3 A's added in a row, there will be 3 units of PPi released.
APS (adenosine 5' phosphosulfate) then reacts with PPi via the enzyme ATP sulfurylase. The ATP sulfurylase reaction forms ATP and sulfate. This reaction is also reversible and may account for the name of the enzyme. I propose here that since there is no ATP in the reaction mixture, the reaction is favored to go forward towards the products.
The ATP then reacts with the substrate luciferin via the enzyme luciferase. This reaction produces AMP, phosphate, oxyluciferin, and light. The light is proportional to the amount of ATP (which correlates to the amount of inorganic phosphate and the amount of base added). By reacting one base at a time, the light (and its intensity) can be correlated back to the identity and quantity of the base.
Posted by: Chad Satori | February 21, 2008 01:20 PM
2) Margulies et. al. stated that the time needed to fragment, prepare, and sequence the genome of 'mycoplasma genitalium' was 4 hours by one individual. I am not aware of how long it takes to do a Sanger sequencing experiment including the time it takes to prepare amplified colonies. Steinberg et. al. suggested that it was longer than the time necessary for the Margulies method.
Posted by: Chad Satori | February 21, 2008 01:25 PM
3) The DNA is loaded into fiber-optic slides. The slides are then loaded into a flow chamber where sequencing reagents can flow above the pores. The chamber delivers reagents (more specifically bases) one at a time. The reagents are able to freely move into the chamber and react (or not react). Margulies reports the time frame of base reaction to be 10 seconds. ATP sulfurylase and luciferase action occurs in 0.02-1.5 seconds. When the correct base is added, the charge-coupled device (located in the closed part of the slide) detects emitted photons. After each reaction of base, apyrase is introduced to the slide to degrade any unused reagents. The slides are created with optic fiber etching.
Posted by: Chad Satori | February 21, 2008 01:34 PM
I have a question that may seem completely dumb, but is there any inherent error to be considered since you are working with such small fiber-optic slides? It just seems like you would introduce significant error when working with this type of medium.
Posted by: Courtney Wettlaufer | February 26, 2008 07:06 PM
In response to Courtney's question, I believe the error can be minimal with the fiber optic slides. The slides are machined by fiber optic etching and the loading and analysis errors can be minimized by automating the process. If the slides were analyzed or loaded by hand, then this may introduce a significant possibility for error that must be accounted for.
Posted by: Josh Ochocki | February 28, 2008 11:16 AM
I think the slide error could be correlated to the decrease in base error frequency. There is only a fixed volume inside the pore for which the bead containing fragmented DNA helicies can be placed. As DNA is extended onto the bead, there might be a limited ability for the DNA to extend OR for the reagents to enter the pore and react. Also if new reagents are not introduced into the system, the Klenow fragments could be degradaded, reagents could become limiting, and ions could be lost in the process.
Posted by: Chad Satori | February 28, 2008 01:33 PM