Elliot Altman (with editing by ECK)


In Sancar's maxicell technique, a recA mutant strain harboring a plasmid is irradiated with a low dosage of ultraviolet light. As a result of the irradiation, the host's chromosomal DNA is destroyed while the plasmid DNA continues to amplify. Thus, with the addition of 35S labeled methionine, only plasmid gene products are visualized.

It was initially thought that a maxicell strain had to be recA and uvrA (the original strain used by Sancar was CSR603). It has since been shown that recA strains also serve well. As a result, many have switched to SE5000 as their preferred strain.

I have tested five maxicell strains and SE5000 is the easiest to work with. It gives good results and, compared to other strains, it is fairly healthy. One of the biggest problems in working with the technique is manipulating the strains: they are very unhealthy because of the recA mutation. This causes two problems:
1. Plasmid transformation efficiency is 2 - 10 fold down from WT.
2. Doubling times are 2 - 4 fold greater than for WT.

This is my technique for generating maxicells from constitutive plasmids such as pBR322. It is a modification of the original method described by Sancar.


1. Transform the plasmids of interest (e.g., pBR322) into the maxicell strains of choice. I would strongly suggest that you use SE5000. if at all possible.

2. The day before you wish to do maxicells, start a 2 mL 37° overnight of SE5000 pBR322 in M9 glucose (recipe below) plus 50 µg/mL Ampicillin.

3. Dilute the 2 mL overnight approximately 1/60 in 30 mL of M9 glucose plus 50 µg/ml Ampicillin. This will yield a 30 mL culture with an OD550 ~0.05. If you are making maxicells from more than one plasmid, adjust all concentrations to about the same OD. Grow the cultures at 37° using vigorous aeration: maxicell strains grow slow enough as it is. I use a New Brunswick shaker bath for this.

4. For each culture, take an initial OD550 reading and readings every hour. Plot the data on semilog graph paper. This will allow you to predict when the culture will be at an OD550 of 0.4.

5. Turn on the UV germicidal lamp 5' before the procedure. When the OD550 nears 0.4, remove 10 mL to a 100-150 mm Petri dish. Leave the top off. The OD550 must be between 0.38 and 0.42. This is critical to insure that the UV dosage per cell is constant between experiments. I work in a hood with the room lights off. Use 90 J/m2 UV while constantly swirling the culture. I achieve this intensity by holding the culture 30" under a UV germicidal lamp mounted 22.5 inches away. Transfer the culture to a 250 mL "Ray-Sorb" dark red flask. The room lights can then be turned back on. Once the maxicells have been irradiated, it is critical that they not come into contact with visible light, which would allow photo-reactivation.

6. Continue incubating at 37° using very slow aeration. Maxicells are beginning to form but it will be at least 15 hours before they can be used.

7. At 30' following irradiation, add cycloserine at 100 µg/mL. This will kill all cells that escaped radiation damage. If left untreated, they would wipe out maxicell expression. I add my cycloserine as follows: Add 0/05 g of cycloserine to 4.95 mL of sterile water. Vortex to dissolve the cycloserine, and filter sterilize using an acrodisc attached to a 10 mL syringe. Add 0.1 mL to the 10 mL culture, giving a final concentreation of 100 µg/mL cycloserine.

Note: Cycloserine is not mandatory, but it enables maxicell formation at much lower UV dosage. 100 µg/ml is the optimum concentration. Since neutral solutions of cycloserine are unstable, I make it up daily and then discard it.

8. Leave the cultures aerating slowly at 37° overnight. They must be left for at least 12 hours and can be left for as many as 18 hours before using them.

(Steps 9-11 are necessary to wash away the cycloserine that is present in the culture.)

9. The next morning, centrifuge the maxicells 6' at 6000x g (7,000 rpm in an SS34). The rest of the procedure can be done under normal room lighting.

10. Resuspend the pellet in 5 mL of M9 glucose + 50 µg/ml Ampicillin.

11. Centrifuge the maxicells 6' at 6000 g.

12. Resuspend the pellet in 1 mL of M9 glucose + 50 µg/ml Ampicillin. Transfer the sample to a 18x 150 mm disposable culture tube. Concentrating the maxicells from 10 to 1 mL makes it easier to add very high concentrations of 35S methionine and facilitates the final analysis by SDS PAGE.

13. Add 50 µCu 35S methionine. Incubate 30' at 37° with moderate aeration. I place the 18x 150 mL culture tubes into a modified test tube rack in a New Brunswick shaker bath.

14. Transfer each sample to a 1.5 mL Eppendorf tube and centrifuge 3' in a tabletop microcentrifuge.

15. Discard the supernatant. I use a Pasteur pipet attached to an aspirator. Warning: The supernatant contains unincorporated 35S methionine which will cause smearing on the autoradiograph if not thoroughly removed.

16. Resuspend the pellet in 200 µL sample buffer for SDS PAGE analysis. I run 10 µL on a 0.75 mm thick gel. Even at this dilution, proteins with a MW < 20,000 will be overloaded. In this case, you will need to do a subsequent 50% dilution. By the time maxicells are harvested, they have an OD550 of ~0.5. That means you have 5 OD550 units protein per sample, but you can only load between 0.2 and 0.4 OD550 units on a 0.75 mm thick gel without overloading the low MW proteins.

HB101 pro, leu, thi, lacY, hsdR, endA, recA, rpsL20, ara-14, galK2, xyl-5, mtl-l, supE44, derived from K12 (TR6710 in Roth collection)

SE5000 F-, amD139, Δlac-169, rpsL, relA, thi, recA56, derived from MC4100

MS3 recA2463, pro+, derived from W3101

CSR603 uvrA6, recA1, phr, derived from AB1886

MS198 Hfr, KL6, recA58, srl-::Tn10, ileu, val, thr, spcR, derived from DF 475 (stabs of material checked for Hfr and recA; transfers recA early anticlockwise)
Most of these strains will produce good maxicell results when used as described above. Since there are slight light variances between different germicidal lamps, a range of dosages should be tested with each new strain. 45, 90, 135, and 180 J/m2 provide a good test range.

Sancar, A. and Rupp, C. "Determination of plasmid molecular weights from ultraviolet sensitivities". Nature, Vol 272., pp 471-472. March 1978.

Sancar, A., Hack, A., and Rupp, D. "Simple method for identification of plasmid-coded proteins" Journal of Bacteriology Vol 137, No. 1, pp 692-693. Jan 1979.

Sancar, A. and Rupp, D. "Physical map of the recA gene". Proceedings of the National Academy of Sciences. Vol 76, No. 7, pp 3144-3148. July 1979.

Roberts, T., Bikel, I., Yokum, R., Livingston, D., and Ptashne, M. "Synthesis of simian virus 40 + antigen in Escherichia coli". Proceedings of the National Academy of Sciences Vol 76, No. 11., pp 5596-5600. Nov. 1979.


M9 Glucose (100 mL)
	80 mL	ddH2O
	10 mL	10x M9
	10 mL	10x Aminoacids
	1 mL	20% w/v glucose
	100 µL	1M MgSO4
	100 µL	thiamine (B1)
	10 µL	CaCl2
10x M9 (1 L)
	60 g	Na2HPO4
	30 g	KH2PO4
	5 g	NaCl
	10 g	NH4Cl
		Bring to 1 L with ddH2O
10x Aminoacids (for low sulphur media; 1 L)
	600 mg	L-Ala
	500 mg	L-Arg
	700 mg	L-Asp
	800 mg	L-Glu 
	300 mg	L-Gly
	200 mg	L-His
	300 mg	L-Ile
	1.1 g	L-Leu
	700 mg	L-Lys-HCl
	300 mg	L-Phe
	300 mg	L-Pro
	300 mg	L-Ser
	300 mg	L-Thr
	400 mg	L-Trp
	200 mg	L-Tyr
	300 mg	L-Val
	Bring to 1 L with ddH2O
	Stir to dissolve
	Filter sterilize.

Last Update: Thursday June 19 2014
This page has been viewed Failed to execute CGI : Win32 Error Code = 2
Eric Kofoid eckofoid at