How
to run an HPLC (or HPLC 101)
High Performance Liquid Chromatography
II)
Prepping
samples
III)
Turning
on HPLC and prepping it
IV)
Troubleshooting
Preparing
buffers is fairly easy as long as you’re not trying to create a new HPLC
Setup. Listed here are the two most common buffers
used in this laboratory.
Buffer
A – Stationary phase
10%
Methanol, 0.01% Acetic Acid, pH 5.05
Add
200 ml HPLC grade methanol
Add
2 ml Acetic Acid.
Qs
to 2L using HPLC grade water.
Adjust
pH to 5.05 using NH4OH.
Degas
for 15 minutes.
This
buffer will be called buffer A in this protocol.
Buffer
B – Mobile Phase
100%
Methanol
2L
of HPLC grade methanol that has been degassed for 15 minutes.
This
buffer will be called buffer B in this protocol.
The samples we will generally use on the
HPLC come from cell cultures, so the very first thing we will need to do is to
centrifuge them (2000 rpm for 10 minutes).
Then
add a known amount of a standard to
every sample. The standard is generally
referred to as the internal control. It should be something that won’t co-elute
with the metabolites from the samples.
Normal standard controls used in this lab include: PGB1 and [3H]
LTB4. (3-6 ul) (NET852).
Save supernatant from each sample by
pouring off into separate 50ml tubes.
To the remaining cell debris add 1ml
Methanol (f.c. = 10%) and 250 ml of Glacial Acetic
Acid. Vortex and transfer to eppendorf
tubes. Sonicate the eppendorf tubes
containing the cell debris for 2 minutes then centrifuge at 14000xg for 10 minutes. Take that supernatant and add it to the
original supernatant. Dispose of the
cell debris in the radioactive waste.
The final volume varies from sample to
sample, but this must remain constant: the final amount of methanol present can
not exceed 10%. For example, with 3 ml
of methanol present you would bring up to 30
ml of 1% Acetic Acid. In all cases,
use 1% acetic acid to bring it up to the final volume. Store on ice until ready to sep-pak.
The sep-pak cartridges are stored in D438
on the top shelf to your left as you immediately walk into the lab. The sep-pak box is stored under the hood of
the room.
The
box should be attached to a waste bottle that is connected to a vacuum
pump. When using this system try not to
exceed 200 psi whenever possible.
Turn on vacuum. Add 5 ml of methanol
and let it wash off. Add 10 ml 0.5% acetic acid (twice
methanol volume) and also let it wash off.
If any methanol is retained on the column still you will lose your
entire sample at this point so be sure you wash it well with acetic acid. Add the samples to each sep-pak cartridge
and allow them to wash through (this is the long step and can take up to 2
hours). When the samples have washed
through, add 5ml of 0.5% acetic acid
and allow that to wash off. After this
final washing step, turn off the vacuum.
Add labeled 15ml eppendorf tubes under each sep-pak
cartridge to collect the next step. Once the tubes are added, add 5 ml of
methanol to each cartridge then turn the vacuum on again. This elution contains all your lipids and
fatty acids.
You can now take the samples and
speedvac/dry them. This normally takes
about 4 hours with 5ml of methanol.
Reconstitute the samples in 180ml methanol and 120ml of 1% acetic acid (300ml final volume).
III. Turning on HPLC and prepping
it
Turn on the following systems, in this order:
both pumps, gradient controller, autosampler, and whichever detectors you plan
on using (the UV detector and radiomatic flo-detector are the only 2 detectors
connected currently).
Before you start anything, make sure you
have enough buffer to run the HPLC for however long you will need it (generally
4L of buffer B and 2L of buffer A will last for ~80 hours). The reservoirs are located on the top shelf
directly behind the pumps. In each
reservoir should be tubing covered by a frit (metal filter). The frit should be completely immersed in
the buffer.
On the bottom part of each pump will be a
tube covered in plastic. This is a draw
tube and lets you rapidly flush each line without having to use the gradient
controller. On each pump, separately,
insert a glass syringe (located next to the pumps), open the draw tube, and
pull slowly on the syringe. This will
let you draw out any air bubbles that may be floating in the tubing. It is imperative that all air bubbles be
drawn out; otherwise they will adversely affect your run. Generally you should draw out a total of 30
ml.
On the autosampler, click on config page
and then scroll down until you get to temp.
Control. Go ahead and turn that on. This will commence a cooling procedure, bringing
the temperature of the autosampler down to 4oC.
Check the ecolume reservoir (behind the
flow scintillation analyzer on the top shelf).
If
that’s not full, go ahead and fill that up (ecolume can be purchased in the
self-service store). Now, once all
that’s done, we’re ready to start.
The gradient controller has a variety of
HPLC methods that we may or may not use.
I’ll
briefly list the three most used (which will be more fully explained on a
separate
Page:
Gradient
1: The lab’s standard HPLC run.
Gradient
3: Lets us detect Prostaglandins quickly.
Gradient
10: Lets us detect HETE’s quickly.
To
start a run, just click on operate gradient then input the correct number. The timed
Event
should always be the same number as the operating gradient. The timed event is what identifies how long
each run should last.
On the autosampler you can now click
autopage. A page will come out that
looks like this:
Step From To #
Inj. Run
Step
just tells you at what part of the setup you’re at.
From
Vial is the first vial the autosampler will inject. The rack inside the autosampler is numbered so it should be
pretty easy to figure out where to start.
To
vial is the last vial the autosampler will inject. The rack inside the autosampler is numbered so it should be
pretty easy to figure out where to end.
#
Inj. is how many times you will inject from one vial. In almost all cases this will be 1.
Inj.
Vol. is how much from each vial you plan on injecting. The autosampler can not inject a volume
greater than 200ml. Normally, we inject half of
every sample when we do our run so, in general, our injection volume is 150ml per run.
Run
Time is how long the autosampler will wait before injecting the next sample.
For
gradient 1 the run time is set at 115 min.
For
gradient 3 the run time is set at 25 min.
For
gradient 10 the run time is set at 50 min.
Double
check these numbers since the times may vary depending on what you’re supposed
to be looking at.
Now it’s time to use the computer. You need to click on the detector icon.
You
should get a message that states that it’s connecting to detector. If you don’t then make sure that the flo-one
is turned on. If you don’t receive the
message and the flo-one is on, then you need to go to user->logoff then log
back on. Once you’re connected, you
will need to choose the correct method to open:
Method->Open->
here will be listed various files.
E ICOH3 is
for detecting just 3H samples
EICOC14 is for detecting just 14C
samples
DUALLAB is for
detecting both 3H & 14C
PGELLEN is used for gradient 3
QUIKSOLV is used for gradient 10
Choose
the appropriate file you’ll want to use.
You can go back and check the files by
Method->Edit-> then check each part of the
file to see if it corresponds with what you want to run. Once done checking go to Run->Setup then check each portion,
including giving a label for your run. Use any system you like as long as it
makes sense, but be careful when you store the data and don’t overwrite. Once
that’s done you want to click on ok then go back to Run->Update Detector. After that you can click on Run->Start.
Now you need to add the samples to the
autosampler vials. The vials are small
and tend to form air bubbles when you add the samples into them, so be sure you
eyeball the vials and flick them if an air
bubble does form. We add half of
whatever we have prepared for the HPLC (normally 150ml).
Also run an extra
sample that contains the same amount of the internal standard of the used for the sep-pak elution and a
known amount of an expected metabolite for the study (this is known as an external standard). This standard run
lets us know the expected elution time of the metabolite of interest, if the
system is working properly, and the base number we can use for recovery (For
recovery, use the peak area obtained from the standard as 100% recovery. The peak areas of the internal standards from the samples should be less, giving
you an indication of how much of the sample was lost during the sep-pak
elution).
Ok, everything should be ready to
run. I like to have a hardcopy of
everything I’ve done so at this point I generally sit down and write out on a
separate sheet of paper what I’m running, the date, the sample identification,
and how they will run on the HPLC.
You
should then click on the autosampler autopage
then start run. The rest of the system is automated and
should run by itself.
Be
sure that you input more samples than you really need to run in the computer as
it only does what the auto sampler sends to it.
Flick
vials to remove air bubbles.
Run
an extra sample with same amount of the known standard an expected metabolite.
Set
auto sampler to 4'C if have many samples.
IV. Troubleshooting
1)
Peaks
aren’t eluting at the same time?
Check to make sure you don’t have any air
bubbles. Check the frits and make sure
they’re completely immersed in buffer; otherwise they’ll be pumping air Check the pH and methanol percentage of your
buffers. Failing all that, check for
leaks in the system.
2)
Low
recovery?
Check to make sure you didn’t have any
methanol retained on the sep-pak. Also,
make sure you added the right amount of internal standard
3)
Peak
are not there.
Check to make sure you have the right detection
systems on. 3H detector will
not spot any 14C labeled sample, no matter how sensitive it is. Alternatively, check for bubbles in the
ecolume line. Bubbles in this line will
do two things. First it will lower the
sensitivity, giving a false low reading.
Second, it will cause 14C samples to bleed heavily into the 3H
channel giving false high data in that detection range.