Hi Christophe,

I also have always used DMEM in 5% CO2 with common cell lines like HeLa Kyoto, HEK293T, etc. But when you make the calculations based on the buffering equations, DMEM in a 5% CO2 atmosphere gives a pH of ~7.6 . I think it's still within the range of physiological pH, which is why there is no problem with the growth of the cells. 10% CO2, however, maintains the pH at ~7.4. So I thought it would be better to set the incubator to 10% CO2 because to my knowledge cells should be grown within pH 7.2-7.4 unless a higher or lower pH is wanted on purpose.

I couldn't find DMEM low glucose with lower NaHCO3 concentration. I found only powder DMEM low glucose where I can add the desired amount of NaHCO3 but I don't wanna deal with its tedious preparation procedure. So I incubated my HeLa cells side by side in DMEM low glucose from Gibco (3.7g/L) in 5% and 10% CO2 atmosphere. In 10% CO2, the color of the medium is exactly as if it's taken from a fresh DMEM bottle (orangish-red, pH 7.0-7.4), while in 5% CO2 the color is dark red (pH 7.6-7.7). This observation is consistent with the calculations.

My colleagues acknowledge now that DMEM requires 10% CO2 but the reason why they don't want to switch to 10% is because they don't want any additional parameters after having done so many experiments in 5% CO2. So I'm not going to try to convince anyone, because it's a matter of keeping all conditions constant throughout their projects. So I wanted to buy DMEM low glucose with lower NaHCO3 for an ideal pH in 5% CO2 but no one sells that stuff (except the powder forms). I think I will just stick to 5% CO2 with the regular DMEM low glucose (3.7g/L).

I just don't understand why all the companies produce DMEM only with 3.7g/L of NaHCO3, as if it's an absolute requirement for DMEM. Thank you for your answer, Christophe.

Dear Mevlüt,

Like your colleagues I have always used media with 3.7g/L of NaHCO3 in a 5%CO2 incubator and it works very well. I have used these media for mouse ES cell culture and differentiation. This is something that people in my field use routinely. I guess that's why your colleagues are not willing to change.

I don't know what type of cells you intend to grow but I could suggest you to compare side by side the DMEM with 1.5g/L and the DMEM with 3.7g/L and see how it affects the behavior of your cells. If you see a major difference between the 2 conditions, it will be easier to convince your colleagues that using DMEM with 5% CO2 is a wrong thing.

Hi Christophe,

I also have always used DMEM in 5% CO2 with common cell lines like HeLa Kyoto, HEK293T, etc. But when you make the calculations based on the buffering equations, DMEM in a 5% CO2 atmosphere gives a pH of ~7.6 . I think it's still within the range of physiological pH, which is why there is no problem with the growth of the cells. 10% CO2, however, maintains the pH at ~7.4. So I thought it would be better to set the incubator to 10% CO2 because to my knowledge cells should be grown within pH 7.2-7.4 unless a higher or lower pH is wanted on purpose.

I couldn't find DMEM low glucose with lower NaHCO3 concentration. I found only powder DMEM low glucose where I can add the desired amount of NaHCO3 but I don't wanna deal with its tedious preparation procedure. So I incubated my HeLa cells side by side in DMEM low glucose from Gibco (3.7g/L) in 5% and 10% CO2 atmosphere. In 10% CO2, the color of the medium is exactly as if it's taken from a fresh DMEM bottle (orangish-red, pH 7.0-7.4), while in 5% CO2 the color is dark red (pH 7.6-7.7). This observation is consistent with the calculations.

My colleagues acknowledge now that DMEM requires 10% CO2 but the reason why they don't want to switch to 10% is because they don't want any additional parameters after having done so many experiments in 5% CO2. So I'm not going to try to convince anyone, because it's a matter of keeping all conditions constant throughout their projects. So I wanted to buy DMEM low glucose with lower NaHCO3 for an ideal pH in 5% CO2 but no one sells that stuff (except the powder forms). I think I will just stick to 5% CO2 with the regular DMEM low glucose (3.7g/L).

I just don't understand why all the companies produce DMEM only with 3.7g/L of NaHCO3, as if it's an absolute requirement for DMEM. Thank you for your answer, Christophe.

I see that you have an eye for the detail. This is a good quality for a PhD student. I wish you good luck for your research project.

Hi Sebastien,

Thanks for your input. I solved my problem by using another incubator with 8% CO2. I'd suggest you to use a different incubator where you can set your desired CO2 concentration because the ATCC DMEM is horribly expensive.

In my original post, I stated that 10% CO2 is required for DMEM (3.7 g/L) to maintain a pH of 7.4. This is actually wrong. When I did the calculations again, I found out that 8.5% CO2 keeps the pH at 7.4. You can calculate the pH of your medium depending on CO2 and NaHCO3 concentrations at 37 degrees Celsius using the following equation:

pH = 6.1 + log10[52 x (NaHCO3(g/L) / %CO2) - 1]

When I open a fresh bottle of DMEM, I take 60mL out and add 50mL of FBS and 10mL of L-glutamine. Therefore the final NaHCO3 concentration is 3.256 g/L. In this case, 8% CO2 maintains the pH at exactly 7.4. 

There are a few things you need to consider, though. If your cell line grows very fast and if you use high glucose DMEM, the pH drop may be significant after 48 hours due to excessive lactic acid build up. Therefore people who culture fast growing cells in high glucose DMEM and leave their cells for more than 2 days in the same medium may prefer 5% CO2 to delay acidification. On the other hand, when you use low glucose DMEM, lactic acid build up never overcomes NaHCO3 buffering capacity even with fast growing cell lines. Therefore 5% CO2 does not bring any benefit for those who use low glucose DMEM.

Best,

Mevlut

" You can calculate the pH of your medium depending on CO2 and NaHCO3 concentrations at 37 degrees Celsius using the following equation:

pH = 6.1 + log10[52 x (NaHCO3(g/L) / %CO2) - 1]"

Can you please explain how you get to that equation?

Dear Mevlüt Çıtır

I am interested also to know how you got the following equation since i am going to work in cell culture ( pH = 6.1 + log10[52 x (NaHCO3(g/L) / %CO2) - 1).

Thanks

Hi, I would like to know how did you get the equation? Would be happy if you could share the reference. Thank you. 

Sorry guys, I didn't check the thread for long time. 

Adrian already shared a reference (I guess that is the only source where you can find the equation in this form) but they don't show how they get to that equation. Actually you don't need a reference, you just need to write down the chemical reactions and calculate the pH. Let me explain how you do that.

In bicarbonate buffering systems, the reactants and products are in equilibrium as shown below:

• CO2 + H2O H2CO3 HCO3- + H+

Let's first focus on the right side of the equilibrium ( H2CO3 HCO3- + H+). The ratio of the forward and reverse rate constants (kf and kr) will give the equilibrium constant (keq), which is also called the acid dissociation constant (ka) of carbonic acid (H2CO3) in this case. In equilibrium, the forward and reverse reaction rates are equal.

• kf * [H2CO3] = kr * [HCO3-] * [H+]
• ka = keq = kf / kr = [HCO3-] * [H+] / [H2CO3]

The negative logarithm of ka is called pKa.

• pKa = -log10([HCO3-] * [H+] / [H2CO3])

            = -log10([H+]) - log10([HCO3-] / [H2CO3])

Because pH = -log10([H+]), we can rearrange the equation as below:

• pH = pKa + log10([HCO3-] / [H2CO3])

This is the famous Henderson–Hasselbalch equation. The pKa of carbonic acid (H2CO3) at 37 degrees Celsius in an aqueous solution that has physiological ionic strength is 6.1. Hence,

• pH = 6.1 + log10([HCO3-] / [H2CO3])

Now we need to rearrange this equation to relate it to the concentration of CO2. Let's focus on the left part of the equilibrium (CO2 + H2O H2CO3). According to Henry's gas solubility law, concentration of CO2 in the liquid phase is a product of partial pressure of CO2 in the gas phase and Henry's constant:

• [CO2(aq)] = pCO2 * k

Henry's constant for CO2 at 37 degrees Celsius in an aqueous solution that has physiological ionic strength is 0.0229 molar/atm. Because our cell culture incubators maintain the partial pressure of CO2 at a constant value, the concentration of H2CO3 is as below:

• [H2CO3] = pCO2 (atm) * 0.0229 (molar/atm)

The partial pressure of CO2 is a product of total pressure and the percentage of CO2 in the gas phase. The total pressure is 1 atm (unless your lab is on a mountain). Hence,

• pCO2 = 1 atm * %CO2 * (1/100)
• [H2CO3] = 1 atm * %CO2 * (1/100) * 0.0229 (molar/atm)

                   = 0.000229 * %CO2 molar

Note: %CO2 represents the number after the percentage (%) sign in this equation.

Now we write this in the Henderson–Hasselbalch equation:

• pH = 6.1 + log10([HCO3-] / (0.000229 * %CO2))

Let's rearrange this equation for our cell culture system. We have an initial amount of sodium bicarbonate (NaHCO3) in our cell culture medium. The concentration of NaHCO3 in the culture medium is also the initial concentration of HCO3- (NaHCO3 --> Na+ + HCO3-). In an equilibrium reaction, if you increase/decrease one of the reactants or products, the system will try to reverse this change until it reaches the equilibrium. So, in our buffering system, having an initial amount of HCO3- will move the reactions to the left (CO2 + H2O  H2CO3 HCO3- + H+). An 'X' amount from the initial amount of HCO3- will be used in the left direction and produce an 'X' amount of H2CO3, which in equilibrium will be equal to the amount of H2CO3 we calculated above by using the percentage of CO2 in the gas phase, which is kept constant by the incubator. So,

• pH = 6.1 + log10(([HCO3-](initial) - X) / (0.000229 * %CO2))
• X = [H2CO3] = 0.000229 * %CO2
• pH = 6.1 + log10(([HCO3-](initial) - (0.000229 * %CO2)) / (0.000229 * %CO2))

           = 6.1 + log10(([HCO3-](initial) / (0.000229 * %CO2)) - 1)

Finally, [HCO3-] is a molar concentration. We want to write this in g/L. Molar mass of sodium bicarbonate (NaHCO3) is 84.007 g/mol.

• [HCO3-](initial) = [NaHCO3] = (g / (g / mol)) / L = (1/84.007) * NaHCO3(g/L) molar

So, the final equation is as below:

• pH = 6.1 + log10(((1 / (84.007 * 0.000229)) * NaHCO3(g/L) / %CO2) - 1)

           = 6.1 + log10((51.98 * NaHCO3(g/L) / %CO2) - 1)

PS: In my previous post, I did not take into account the amount of sodium bicarbonate present in FBS when calculating the final concentration of sodium bicarbonate in the culture medium upon addition of FBS and L-glutamine. FBS contains around 2 g/L of sodium bicarbonate. Check the lot specific certificate of analysis of your serum to find the exact value, if possible.

Wow, you write up all of these explanation? Such a great explanation! Thanks a lot!

You're welcome, Khairunnisa. Since several people asked for it, I wanted to explain it in great detail.

Cheers!

Sorry to add something at this point, but the pH of tissues is about 7.15-7.20. While it's true that blood plasma has a pH of 7.35-7.45 (as a clinically trained MD, I know this to be true), this is not the case in tissues - WHICH IS WHERE THE CELLS LIVE (http://www.finom.fi/syopajahappamuusihmisessa.pdf ).

DMEM with 3.7 g/L of bicarbonate was designed by two Nobel prize winners (Harry Eagle and Renato Dulbecco) to be used with 10% CO2 in which case the media has a distinct rust color (not the purplish red you see at 5% CO2). In contrast, the ATCC sells a DMEM with 1.5 g/L of bicarbonate which is to be used in a 5% CO2 incubator (https://www.atcc.org/products/all/30-2002.aspx ). If you've been doing it wrong and succeeding your whole life, why stop now! On the other hand, if you're an old dog who can in fact learn new tricks, then why not consider making a change. In our lab, we have two incubators: (1) 5% CO2 for RPMI 1640 and (2) 8% CO2 for DMEM and FreeStyle media from GIBCO (which recommend 8% CO2).

(And then there's always the question of getting a 3 gas incubator for 5% O2 which again is more like the concentration of oxygen in tissues.)

Hi Mevlut,

Thank you for the detailed explanation. There is one step I am not following:

"An 'X' amount from the initial amount of HCO3- will be used in the left direction and produce an 'X' amount of H2CO3, which in equilibrium will be equal to the amount of H2CO3 we calculated above by using the percentage of CO2 in the gas phase, which is kept constant by the incubator."

This seems to assume that all of the H2CO3 in solution comes from HCO3- and none of it comes from atmospheric CO2. Why can we make this assumption here? Is this because the reaction CO2+H2O H2CO3 is much slower?

Instead, should we consider that [Na+] + [H+] = [HCO3-] due to charge neutrality? Assuming [H+] 6.5 (L pKa is 3.86). Therefore, [HCO3-] = [Na+] - [L-], independent of %CO2. This shows that a higher amount of NaHCO3 provides buffering up to high levels of lactic acid.

Hi,

I found this highly valuable disscusion by chance and I must admitt I also have been routinely using DMEM in 5% CO2 without thinking about it... Just to be precise: H. Eagle was awarded the National Medal of Science, but not a Nobel Prize (https://www.nytimes.com/1992/06/14/nyregion/dr-harry-eagle-is-dead-at-86-formulated-cell-growth-medium.html). Thanks to all!

As I understand it, DMEM is a formula. It has certain quantities of various ingredients, and if you differ from that formula in your manufacturing recipe, you can't continue to call it DMEM. A manufacturer can maybe call it "DMEM buffered with HEPES instead of Sodium Bicarbonate" but then you have to use that name in all of your publications involving that reagent. Since most folks are doing science connected with prior publications, there's a trend to want to use the same reagents used "as published in [wherever]" which would not have that modification. Even if it's not the optimal choice for the actual health of the cells.

So basically, my answer is "tradition".

For thousands of years, it was also a tradition to say that the Earth was flat.

When writing a paper, you should say that you used DMEM and name the supplier. For DMEM from GIBCO, GIBCO has its formulations listed on their website and it will say "3.7 g/L of bicarbonate." For DMEM from the ATCC, it will say 1.5 g/L of bicarbonate and they note that it is intended to be used in a 5% CO2 incubator. So specifying the manufacturer can help eliminate the confusion.

That said, most people - WRONGLY - use 5% CO2 incubators with DMEM from GIBCO. But it generally doesn't make much difference to the experiments - unless you find that your cells don't grow so well. In that case, you need to up your game somehow.

I recommend that people continue to use their 5% CO2 incubators but just buy DMEM from the ATCC, which is formulated for 5% CO2. That way, nobody gets mad and the cells are very happy.

I think Mevlut Citir’s answer is essential. However, it is theoretic. There are many constiuents in media that influence pH. Media formulators use sodium bicarbonate to some degree as a replacement for sodium hydroxide. The amount of bicarbonate in a medium is not affected by how much sodium bicarbonate added. Only the pH and CO2 atmosphere influence that. I use no sodium bicarbonate and culture under 7% CO2. I bubble the gas mixture into my medium, add an amount of sodium hydroxide short of the molar amount calculated by the equation, so substituting sodium bicarbonate with NaOH, equal molar, and further adjust to pH 7.3. Once I do this for a formulation I can simply add this amount the next time if all constituents are the same. If you add a tiny amount of phenol red, the pH is perfect when you argue with yourself about whether the color is orange or pink.

This is quite old discussion, but just noticed it so I have to comment one thing.

Mevlut Citir had very nice answer with all the equations, I wish I would have read it earlier! :)

About the pKa of carbonic acid: in Mevlut's answer and in the reference that Adrian Mateusz Kasztelan provided, pKa was 6.1

However, I have noticed slightly larger reported pKa of H2CO3 (6.2-6.38) at 37C:

6.38 [1], 6.3 [2], 6.2-6.3 [3]

Just thinking, as this pKa basically has direct impact when calculating pH value --> if pKa is 0.2 larger (6.1->6.3), also pH would be 0.2 larger.

... at least I think so. My experts are far away from the chemistry, so it might be that I have missunderstood something.

(Some references are not the best: just have read them through Google books)

[1] M. J. Gramer and T. Ogorzalek, “A semi-empirical mathematical model useful for describing the relationship between carbon dioxide, pH, lactate and base in a bicarbonate-buffered cell-culture process,” Biotechnol. Appl. Biochem., vol. 47, no. 4, pp. 197–204, 2007.

available: http://www.chembio.umich.edu/publications/ogorzalek.biotechnol%20appl%20biochem.2007.pdf

(for pKa, it's citing to [25] which I do not have access:

[25] Glacken, M.W. (1991) in Animal Cell Bioreactors (Ho, C.S. and Wang,D.I.C., eds), pp. 373–404, Butterworth-Heinemann, Boston, MA )

[2] Tim Smith, Colin Pinnock, Ted Lin, Robert Jones. Fundamentals of Anaesthesia. 2009. p. 351

available: https://books.google.fi/books?id=jSFXYj3ymugC&pg=PA351&lpg=PA351&dq=bicarbonate+buffer+system+pK&source=bl&ots=25W_yZV-V6&sig=c12spQh3F_S-qIO44zAUercBHFg&hl=fi&sa=X&ved=0ahUKEwjDwvecksfLAhXoFZoKHRZFDbIQ6AEIUTAJ#v=onepage&q=bicarbonate%20buffer%20system%20pK&f=false

[3] Lubiniecki, Large-Scale Mammalian Cell Culture Technology, 1990, p. 462

available: https://books.google.fi/books?id=wrjld3en-p0C&pg=PA464&dq=NaHCO3+cell+culture&hl=fi&sa=X&ved=0ahUKEwiLtuz3wcfLAhXMCpoKHZ3qA7QQ6AEIQDAC#v=onepage&q=pKa&f=false

I have changed CO2% of the incubator until the pH of DMEM was 7.4 and the concentration of CO2 for that pH was 9%. I buy the powder and I am tempted to add 1.5 of NaHCO3 (more physiological) as ATCC media (I use gibco) but ATCC only comes with high glucose and I am using low glucose. Would that be correct?

Mariana, If you're using DMEM from GIBCO/LifeTech or any other supplier than ATCC, then the correct CO2 concentration in the incubator would be about 10% - NOT 5% as many people mistakenly use. It's interesting that you tested for pH 7.4 and found that 9% CO2 is good. If your lab colleagues do not want to give you a whole incubator for your cultures that are not at 5% CO2, you can try the ATCC media which is fine at 5% CO2 or, as you say, use powdered DMEM to which you add 1.5 g/L NaHCO3. If you do this, I recommend you use Baxter Water for Irrigation (medical grade low endotoxin water). If you use Millipure or other types of water, you could accidentally have serious endotoxin contamination - this is especially bad when working with monocytes and macrophages:

Article Tumor necrosis factor production by human monocytes is a reg...