Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Google      Sign In   
  • Create Account

Submit your paper to J Biol Methods today!
Photo
- - - - -

Failed cloning - different readings


  • Please log in to reply
10 replies to this topic

#1 Biologystudent

Biologystudent

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 32 posts
0
Neutral

Posted 21 March 2014 - 04:30 AM

I am to clone a gene into a vector.

 

1) I mini prepped the gene to be inserted -----> ran a PCR with primers flanking the gene  -----> digested the ends of the PCR-product to create compatible ends

 

2) I ran a electrophoresis of the vector... and got 1 single intact band.

I then digested the vector. and electrophoresis showed 2 separate bands (bigger band was cut vector and smaller band was the excised unwanted fragment). I cut out the vector from the gel and purified it

 

3) I ligated the vector and insert. 1:3 ratio gave me 13.5ng insert to 45ng vector. Which based on their separate concentrations are equivalent to close to 1mikroliter insert + 2mikroliter vector.

 

4) I transformed bacteria with this ligation-sample. But I got no growth.

 

5) I ran a electrophoresis of my vector backbone and it showed NO BANDS at all. So assumably, I lost the vector during the purification. Right? Is that a correct interpretation?

 

6) I also ran a electrophoresis of the insert and it showed 2 bands when I was expecting 1 band. So assumably, the insert had been cleaved. Right? Is that a correct interpretation? But how? Where and how did this unwanted cleavage occur?

 

7) I measured the vector again after all this and the nano-drop now gave me a reading that was TWICE as high as before when I did the experiment. WHAT?????!!!

 

Can someone please ease my mind and tell me what they think of all this?


Edited by Biologystudent, 21 March 2014 - 04:31 AM.


#2 phage434

phage434

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 2,440 posts
241
Excellent

Posted 21 March 2014 - 07:11 PM

1) did you look at the PCR product on a gel to determine if the PCR worked? Did you purify your PCR product before cutting it? Did you look at the cut PCR product on a gel?

If the cut fragment has two bands, likely there is a restriction cut site in your pcr fragment.  Did you include some junk DNA bases (4-6 bp) 5' of your restriction site on your primers?

Did you either heat kill your enzymes, or purify the fragment after cutting?

 

2) Did you verify that the cut fragment was still present after gel purification? Probably it is gone, given your gel results. If you are using different restriction enzymes on the two ends of your insert, then the gel purification may be optional. I would try to eliminate it.

 

3) 4) 5) 6) all likely wasted effort. Almost all "ligation" failures are poor quality or missing DNA going into the reaction. Make sure your fragments are good before ligation.

 

7) Low concentration values on the nanodrop cannot be trusted. It's easy to go from 3 ng/ul to 6 ng/ul when there is 0 ng/ul of DNA.

A gel typically does not lie, and will show you around 20 ng of DNA in the amount of sample you load (2 ng/ul if you load 10 ul).



#3 Biologystudent

Biologystudent

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 32 posts
0
Neutral

Posted 23 March 2014 - 03:44 AM

1) I looked at the PCR on a gel-electrophoresis before I proceeded with the digestion. As a quality-control for the PCR I had 2 dilutions (1:100 and 1:200) plus a no-DNA control. On the electrophoresis the control was good, but I only got a band on the 1:200 dilution-sample. So I proceeded my work with this sample only.

I purified this 1:200 sample after the PCR, then digested it and then purified it again (before the final ligation).

Yeah, had 6bases of As.

Digestion was for 15min at 37degrees and no heat inactivation was done (not needed according to protocol and supervisor. Actually, supervisor was STRICTLY AGAINST any inactivation-step since that might affect the samples themselves). But I did purify the sample once again after the digestion.

 

 

2) For the vector itself, I did not run a electrophoresis after the purification (before the ligation), to see whether they were still there. But I did it once the experiment was done and failed, and then I saw absolutely NOTHING/NO BANDS at all.

 

 

3/4/5/6) How common is this? I feel completely retarded and that I am wasting my supervisors time and interest.

 

We usually load our gels with 1-5mikroltiers of sample + 2mikroltiers loading dye + and remaining volume with dH2O up to a total volume of 12mikroliters.

 

 

 

1) did you look at the PCR product on a gel to determine if the PCR worked? Did you purify your PCR product before cutting it? Did you look at the cut PCR product on a gel?

If the cut fragment has two bands, likely there is a restriction cut site in your pcr fragment.  Did you include some junk DNA bases (4-6 bp) 5' of your restriction site on your primers?

Did you either heat kill your enzymes, or purify the fragment after cutting?

 

2) Did you verify that the cut fragment was still present after gel purification? Probably it is gone, given your gel results. If you are using different restriction enzymes on the two ends of your insert, then the gel purification may be optional. I would try to eliminate it.

 

3) 4) 5) 6) all likely wasted effort. Almost all "ligation" failures are poor quality or missing DNA going into the reaction. Make sure your fragments are good before ligation.

 

7) Low concentration values on the nanodrop cannot be trusted. It's easy to go from 3 ng/ul to 6 ng/ul when there is 0 ng/ul of DNA.

A gel typically does not lie, and will show you around 20 ng of DNA in the amount of sample you load (2 ng/ul if you load 10 ul).



#4 phage434

phage434

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 2,440 posts
241
Excellent

Posted 23 March 2014 - 05:23 AM

Well, it sounds as if many things are working and were done correctly, so that's good. You major difficulty is that you lost all of your DNA when you gel purified your vector. I hate gel purification, so I'm a poor person to tell you details of how to do it. Are you using two different enzymes for the two ends of your insert, or a single enzyme? With different ones, you may be able to simply purify your vector DNA after cutting and go. If there is a single enzyme, you should be able to dephosphorylate your vector DNA. Use an enzyme such as shrimp alkaline phosphorylase that can be heat killed, and do not use an excess or use it for long periods. You should run the "no insert" control when doing your ligations/transformations, which will tell you if you have uncut  vector or religated vector (and roughly how much).



#5 Biologystudent

Biologystudent

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 32 posts
0
Neutral

Posted 24 March 2014 - 01:31 AM

Did I somehow get DNase into my samples at some step?

2 different enzymes:

- insert is cut with RE1 at 5' and RE2 at 3'

- vector is cut with RE3 at 5' and RE2 at 3'

 

RE1 and RE3 give compatible ends (reason we use RE3 instead of RE1 for the vector is because the insert also has a site for RE3 in the middle of the gene, and thus would ruin everything).

 

I do just purify my vector after digestion, and then it is ready for the ligation.

We don't work with dephosphorylation. What would that do? Is it to prevent self-ligation? In any case, we lost the DNA, so that wouldn't even be happening.

 

I also had a control-ligation-transformation (vector alone+ligase control). Didn't get any growth on that plate either (as expected though). So what can I interpret from that?

 

Well, it sounds as if many things are working and were done correctly, so that's good. You major difficulty is that you lost all of your DNA when you gel purified your vector. I hate gel purification, so I'm a poor person to tell you details of how to do it. Are you using two different enzymes for the two ends of your insert, or a single enzyme? With different ones, you may be able to simply purify your vector DNA after cutting and go. If there is a single enzyme, you should be able to dephosphorylate your vector DNA. Use an enzyme such as shrimp alkaline phosphorylase that can be heat killed, and do not use an excess or use it for long periods. You should run the "no insert" control when doing your ligations/transformations, which will tell you if you have uncut  vector or religated vector (and roughly how much).



#6 phage434

phage434

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 2,440 posts
241
Excellent

Posted 24 March 2014 - 08:55 AM

If you didn't get any growth on your insert-negative control, then I would assume your competent cells are not very good. There is almost always uncut plasmid or religated plasmid which can transform. Test your competent cells by doing serial dilutions of your uncut plasmid to 10 pg/ul and transforming a microliter. You should get tens to hundreds of colonies.



#7 Biologystudent

Biologystudent

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 32 posts
0
Neutral

Posted 25 March 2014 - 01:01 AM

The cells we use are "ultra competent" according to manufacture. We also did the transformation again using rich SOC-media. Still no growth.

What do you mean by "Test your competent cells by doing serial dilutions of your uncut plasmid to 10 pg/ul and transforming a microliter"

I am sorry for not understanding the first time. All of this is new to me. And English isn't my first language either.

 

Question 1)

Do you mean that I should take my plasmid that carries the vector+insert, measure the DNA-concentration and accordingly make a dilution-series ranging from start-concentration down to 10pg/ul. And then do transformations with all of these dilution-samples? And tomorrow check for growth on the plates.

 

Question 2)

Or should I just use my plasmid carrying the vector itself (no insert, no digestion, no nothing). If this is what you mean, then how would this be any different from the plate that I already have with transformed cells carrying my vector. The miniprepp I start out with (and then do all of the other steps from: PCR, electrophoresis, digestion, ligation with insert), originates already from a plate with transformed bacterias that have taken up the vector.

 

If you didn't get any growth on your insert-negative control, then I would assume your competent cells are not very good. There is almost always uncut plasmid or religated plasmid which can transform. Test your competent cells by doing serial dilutions of your uncut plasmid to 10 pg/ul and transforming a microliter. You should get tens to hundreds of colonies.



#8 phage434

phage434

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 2,440 posts
241
Excellent

Posted 25 March 2014 - 10:00 AM

I meant (2). Establishing the competence of your cells and transformation techniques is very important, because ligation, even under good circumstances, produces low numbers of correct circular molecules. Competence is usually expressed in colony forming units per microgram of DNA. If you dilute your uncut vector to concentration of 10 pg/ul (10^-5 ug per microliter) and transform with 1 ul, you can determine your efficiency.  Good cells and technique will give 10^8 to 10^10 cfu/ug, or 10^3 cfu to 10^5 cfu in the transformation I suggested. If you are getting less than 100 in that test, you will have trouble with ligation reactions no matter what else you do. People often assume that because they can transform prepared miniprep plasmid that their cells are sufficiently competent. This is definitely NOT true, and is a leading cause of problems.



#9 lucilius

lucilius

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 109 posts
2
Neutral

Posted 04 April 2014 - 09:51 AM

I meant (2). Establishing the competence of your cells and transformation techniques is very important, because ligation, even under good circumstances, produces low numbers of correct circular molecules. Competence is usually expressed in colony forming units per microgram of DNA. If you dilute your uncut vector to concentration of 10 pg/ul (10^-5 ug per microliter) and transform with 1 ul, you can determine your efficiency.  Good cells and technique will give 10^8 to 10^10 cfu/ug, or 10^3 cfu to 10^5 cfu in the transformation I suggested. If you are getting less than 100 in that test, you will have trouble with ligation reactions no matter what else you do. People often assume that because they can transform prepared miniprep plasmid that their cells are sufficiently competent. This is definitely NOT true, and is a leading cause of problems.

Just wondering: do you always check this competence in this way?

I make a lot of competent cells (I really use lots of them) and it seems pretty hard (a lot of time lost) to check this each time?



#10 Trof

Trof

    Brain on a stick

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 1,199 posts
109
Excellent

Posted 04 April 2014 - 10:49 AM

I don't check competency of purchased cells when they are stored right and are don't due their "best before" if I don't have transformation problems.

 

But I would make transformation control for home-made cells each time I'm using them. It's not that much time, you just plate one more aliquote on one more plate. If you are using your cells continuously you will be able to see decrease in transformation efficiency nearing the limit before you lose ligation reaction on it.


Our country has a serious deficiency in lighthouses. I assume the main reason is that we have no sea.

I never trust anything that can't be doubted.

'Normal' is a dryer setting. - Elizabeth Moon


#11 phage434

phage434

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 2,440 posts
241
Excellent

Posted 04 April 2014 - 06:30 PM

Just as Trof says. You do this when you have doubts about the cells. Newly purchased commercial cells are usually just fine, but home made ones vary by 3-4 orders of magnitude in their competence, and some don't work at all. When things don't work, you need to do controls, even for things you know "always" work.






Home - About - Terms of Service - Privacy - Contact Us

©1999-2013 Protocol Online, All rights reserved.