A convenient apparatus and tracking dye for anaerobic analytical polyacrylamide-gel electrophoresis

ANALYTICAL

BIOCHEMISTRY

83, 337-345 (1977)

A’Convenient Apparatus and Tracking Dye for Anaerobic Analytical Polyacrylamide-Gel Electrophoresis CHARLES

E. MCKENNA, HOP T. NGUYEN, CANDICE AND STEVEN A. SMITH

C~emicaf-Ecological Soufhern

development Laboruro~, Califarnia, University Park,

department Los Angeles,

of Chemists,

California

W. HUANG, University 90007

of

Received April 5, 1976; accepted July 19, 1977 An anaerobic modification of conventional polyacrylamide-gel electrophoretic equipment is described. The modified apparatus has been applied to the separation of Azotobacter vinelandii nitrogenase components and should prove useful in the analysis of other O,-sensitive proteins. Electrophoresis in reducing gels can be followed with a dithionite-resistant tracking dye, potassium guaiazulene-I-sulfonate.

Electrophoresis in cylinders or slabs of polyacrylamide gel has undergone wide development in the study of proteins, embracing the analytical separation of complex mixtures; measurements of molecular weight, radius, free mobility and relative charge; and the detection of subunits (la-d). Several types of apparatus are offered commercially, but equipment specifically designed for gel electrophoresis in a nonaerobic environment, for applications involving O,-labile proteins or the use of specific gas mixtures, has not been available. We have devised a simple modification that converts a standard commercial gel electrophoresis unit to anaerobic use. The modified apparatus has been successfully evaluated for polyacrylamide-gel electrophoresis of the O,-sensitive, essential protein components of nitrogenase from Azotobacter vinefandii. A possible solution to the problem of dye tracking in strongly reducing gels is also presented. MATERIALS

AND METHODS

Anaerobic experiments were performed under chromous sulfatescrubbed Ar or under H, passed through a Deoxo purifier (Englehardt). Nitrogenase from Azotobacter vinelundii OP (ATCC 13705) was prepared and assayed by modifications (2) of published procedures. Guaiazulene (1,4-dimethyl-7-isopropylazulene), I, was prepared from technical guaiene by an elementary procedure (3). Both compounds are available from Aldrich Chemical Co. Guaiazulene - 3 - sulfonic acid (5 - isopropyl - 3,8 - dimethyl - 1 -azulenesulfonic acid), II, was prepared from I and purified as the potassium salt, 337 Copyright Q 1977 by Academic Press, Inc. All rights of reproduction in any form reserved.

1SSN 0003-2697

338

MCKENNA

ET AL.

III (monohydrate),

according to the method of Meier er al. (4). Spectral constants (‘H nmr, uv-visible absorption) of the recrystallized salt were in accord with the published values (4). Construction

of the Apparatus

The apparatus was based on a Hoefer Model DE 102 polyacrylamidegel electrophoresis unit in which the lid supplied by the manufacturer was replaced by the perforated closure piece shown in Fig. 1. Into the base of the unit were machined six ‘Win.-deep holes threaded to accept a lo-32 bolt and disposed to correspond to the six small holes drilled along the periphery of the closure piece (Fig. 1). Into each base hole was tightly screwed a 9’16 x 81%in. stainless steel rod having %-in. of lo-32 thread at both ends. The apparatus was assembled as follows: First, a well-greased 5%in.-i.d., r/s-in. O-ring was placed between the juncture faces of the lower and upper buffer chambers. A well-greased l%ti-in.-i.d., %-in. O-ring was then fitted into the internal groove of the closure piece collar, and a 5%in.-i.d., l/s-in. O-ring was seated along the large groove circumscribing the lower face of the closure piece, which was then fitted over the upright base rods and pushed down against the lip of the upper buffer reservoir. A vacuum- and A-

FIG. 1. Diagram of closure piece for anaerobic gel electrophoresis apparatus. The piece was made from a %-in-thick slab of Plexiglas, cut into the hexagonal pattern shown to conform to the shape of the base of the Hoefer DE 102 unit. The sides are 2% and 5%~ in. long: These dimensions are not critical. Placement of the six mutually equidistant, doubly chamfered 0.0575-in. holes at the radius indicated provides a slight offset relative to the gel tube holes in the unmodified apparatus; this may be eliminated, if desired, by use of a 17%-in. radius. The O-ring groove around the wall of the central hole should be so situated that the O-ring bears tightly against the safety lock pillar of the upper buffer chamber.

ANAEROBIC

GEL

339

ELECTROPHORESIS

pressure-tight seal was ensured by fastening the closure piece with wing nuts bearing against %6-in. brass washers: The completely assembled apparatus is illustrated in Fig. 2. Polyacrylamide

Gels

All gel reagents were obtained from Bio-Rad, with the exception of buffer components which were obtained from Matheson, Coleman and Bell (glycine) or Sigma. The three different gel systems (A, B, and C) used to evaluate the apparatus in the electrophoresis of nitrogenase are described in Table 1. Small-pore gels were cast with the aid of a Bio-Rad

FIG. 2. The assembled lower buffer chambers

apparatus. The O-ring is not shown.

that provides

a seal between

the upper

and

340

MCKENNA

ETAL.

TABLE1 POLYACRYLAMIDE-GEL

SYSTEMS

FOR ANAEROBIC

ELECTROPHORESISOFNITROGENASE

A

2so

0.625

0.06 M Tris-HCI, pH 6.7

7

0.184

0.375 M TrisHCI, pH 8.9

0.3S4 t.4 GlycineTris (0.05 M), pH 8.3

(3

B

2.50

0.625

0.06 M ImidazoleHCI, pH 5.7

6

0.200

0.06 M TrisHCI, pH 7.9

0.034 M AsparagitteTris. pH 7.3

(lb)

C

4.0

0. I

0.375 M Tti-HCI, pH 9.0

S

0.200

0.375 M TrisHCI, pH 9.0

0,065 M Borate-Tris, pH 9.0

(‘3

* Sample diiuent, 0.@5 M imidazole-HCl

in SO’%glycerol,

pH 5.7, for system B

Model 210 gel preparation system: I .6 ml of small-pore solution and 0.2 ml of large-pore solution were used in each 75 x j-mm (i.d.) glass gel tube. The mounted gels were seasoned overnight in a humidor.

The integral upper buffer chamber-gel holder was first installed on the lower buffer chamber (containing a magnetic stir bar), with a sealing O-ring between. The lower buffer chamber was then filled with 1000 ml of buffer, and the tubes of freshly prepared gel were inserted (positions i-5,711) and secured with rubber grommets in the usual manner. Positions 4 and 12 were occupied by empty tubes (standard 7.5 x 5-mm tubes will serve, but 50 x 5-mm tubes are somewhat more convenient) having their lower ends placed midway between the ceiling of the lower buffer compartment and the surface of the contained buffer. The closure piece was next fitted loosely over the upper buffer compartment as described above, rotated to locate the 0..575-in. perorations midway between the upper ends of the gel tubes, and tightly fastened. Each perforation was plugged with a truncated 15-mm sleeve-type septum stopper (Scientific Products R-5205). The apparatus was connected to a vacuum pump and Ar source via an 18-gauge needle and two-way stopcock, alternately evacuated three times, then filled with Ar; excess pressure was vented, and 10 ml of 0.08 M DT (Na&OJ were injected into the stirred lower buffer via one of the inlet tubes. The samples, consisting of a 50-~1 ahquot of 5- 100 ~1 of protein solution dimted anaerobically to 200 ~1 with 40% sucrose or 50% glycerol, were layered on the large-pore gels, care being taken not to touch any liquid to the inner tube walls. Tracking dye IKI was applied to gels directly (IO-50 ~1 of a

ANAEROBIC

GEL ELECTROPHORESIS

341

2 x 10P4 M solution in 30% sucrose-upper buffer) or dissolved in upper buffer to a concentration of 2-9 x 10e7 M. Each tube was then filled to the top (convex miniscus) with upper buffer that had been degassed three times and stirred with At-, then made 0.0008 M in DT. The remainder of this buffer (ca. 300 ml total) was then transferred under Ar pressure into the upper reservoir, using a l&gauge double-tipped stainless steel needIe (Aldrich Chemical Co.). After the electrophoresis run, gels were fixed and stained with 1% amido black in 7% acetic acid for 5 min, then diffusion destained in 7% acetic acid (O-40% methanol) overnight (Bio-Rad Model 172 diffusion destainer), and stored in 7% acetic acid. Stained gel scans were carried out at 600 nm in a Beckman ACTA VI M spectrophotometer.

FIG. 3. Bromopbenol blue (bpb) and III as tracking dyes in aerobic and reducing (1 mM dithionite) anaerobic polyacrylamide gels. Gel system: A (Table I); electrophoresis conditions: 185 V, 5 mA/gel for 1 hr. Photograph taken -2 min after removal of gels from electrophoresis apparatus. Legend: A = 500 nmol III. aerobic: B = 16.8 nmol bpb, aerobic: C = 500 nmol III + 16.8 nmol bpb. aerobic; D = 16.8 nmol bpb, anaerobic; E = 500 nmol III, anaerobic. Dye loads were chosen to give approximately equivalent visibility with an amount of bpb -2x larger than the normal load of III.

342

MCKENNA

RESULTS

ET AL.

AND DISCUSSION

Anaerobic gel electrophoresis broadly implies absence of any strong oxidant, including 02, from samples and gels. The simple, nondestructive modification described here converts an ordinary electrophoresis cell into a gas-tight chamber with provision for evacuation and filling with inert gas, for buffer addition, and for sample introduction. As a precaution, dithionite is added to the electrophoresis buffer (7); it has been assumed that when the electric field is applied, the small S,0,2- ion migrates with or ahead of the fastest-moving protein, thus scavenging 0, and other immobile residual oxidants in the gel.’ As Fig. 3 shows, when millimolar concentrations of dithionite are present in the upper buffer, reduced bromophenol blue does not reoxidize during anaerobic electrophoresis into a persulfate-initiated polyacrylamide gel. A practical consequence of the preceding observation is that bromophenol blue cannot be used as a tracking dye in anaerobic gels made reducing with dithionite.2 In seeking a satisfactory substitute, we considered several critiera: (a) resistance to bleaching by 1 mM dithionite; (b) minimal molecular size; (c) absence of transition metals; (d) ready visibility (blue or other “dark” color); (e) chemical inertness to proteins and stability to storage; and (f) availability. Generically, azo, azine, indigoid, and thiazine dyes were ruled out as too easily reduced by dithionite to leuco forms. After examination of several other possibilities, we concluded that the azufene chromophore might satisfy criteria (a)-(f). Transfer of a dithionite electron to the antibonding LUMO of this neutral aromatic hydrocarbon system [E1,20 (dioxane-H,O) = -1.6 V] (10,ll) should be considerably more difficult than electron transfer to the LUMO of most conjugated heteroatom dyes. In practice, we elected to study guaiazulene, I, in preference to the more costly parent compound: the effect of alkyl substitution on azulene electron affinity, although small (lo), also argues in favor of I.

I

II

m

1 Dirksen and Chrambach have reported that gels initiated by either persulfate or illuminated riboflavin oxidize sulfhydryl groups. These authors used thioglycolate to reduce their gels (8). 2 Recently this difficulty was explicitly noted, but left unresolved in a paper by Starchenkov and Zhelyuk (9).

ANAEROBIC

GEL

343

ELECTROPHORESIS

Sulfonation is a simple and obvious route to a water-soluble, anionic guaiazulene-based tracking dye. Guaiazulene may be sulfonated with oleum to give a mixture of guaiazulene-2-sulfonic acid, IV, and (Cmethyl7-isopropyl-1-azulyl)-methanesulfonic acid, V, or with acetic anhydridesulfuric acid to guaiazulene-3-sulfonic acid, II (4). The latter synthesis has the advantage that a single product is formed, and in 3.5x higher yield than either isomeric sulfonic acid made in the oleum reaction. Differences among the l Visvalues for isomers II-V are not large enough (4) to affect this choice. The potassium salt of II, III is stable in aqueous solution and can be stored (tightly stoppered) in the dark for months without significant decomposition. As the intensity of visible absorption by III (+&H20) = 640) is two orders of magnitude smaller than that of bromophenol blue (%00 = 6.61 x 104), to achieve comparable visibility a more concentrated solution of III must be used. Figure 3 shows results for gel electrophoresis of bromophenol blue and III under aerobic and anaerobic conditions. In aerobic gels, R,values of the two dyes were not distinguishable within experimental error. In anaerobic dithionite-treated gels, III retained good visibility while bromophenol blue, as expected, was visually undetectable. In normal analytical runs lasting l-2 hr no decrease in the intensity of III could be perceived as it progressed with dithionite through the gels described in Table 1. After incubation of 3.5 x lop5 M III with 10P3M dithionite (pH 7) for 5.5 hr in a sealed cuvette, the decrease in absorbance at 370 nm was less than 5%,3 suggesting that III could also be used to follow relatively protracted electrophoretic separations in dithionite-treated gels. The analysis of nitrogenase fractions is an important application of anaerobic gel electrophoresis. The two metalloprotein components of the enzyme (Fe-MO and Fe proteins) are rapidly and irreversibly inactivated when exposed to air, with the Fe protein showing particular sensitivity. Oxygen-damaged A. vinelandii Fe protein was reported to give multiple banding in an 8% polyacrylamide gel at pH 9.0, while undamaged protein migrated as a single sharp band (12); active or O,-deactivated Klebsiella pneumoniac Fe protein reportedly gave multiple bands in a 6% gel at pH 7.9; mersalyl pretreatment gave a single band (13). In the same gels, the respective Fe-MO proteins gave single bands, and mersalyl-treated Fe-MO protein from soybean nodule bacteroids was reported to migrate as a single band in the 6% gel. (14). We have evaluated the gas-tight electrophoresis apparatus, with III as a tracking dye, for analysis of A. vinelandii nitrogenase components with three gel systems (Table 1). Under anaerobic conditions, the isolated Fe and Fe-MO proteins moved as single bands. Spectrodensitometric scans of aerobic and anaerobic Fe protein gels indicated that 75% or more of the Fe protein band intensity was lost in aerobic runs, 3At the end of the experiment excess dithionite concentration titration with a deoxygenated solution of bromophenol blue.

was confirmed

by back-

344

MCKENNA TABLE R, VALUES

FOR A.

VINELANDII

ET AL. 2 FE-MO

AND

Gel system (Table 1)

Fe-MO protein

A B C

0.52 0.27

FE PROTEINS”

0.41

Fe protein 1.0 1.0 0.86

D R, is defined as R, relative to the guaiazulene dye III.

while in aerobic Fe-MO protein gels the band intensity decrease was considerably smaller. R, (R,relative to III) values for each Azotobacter nitrogenase component are recorded in Table 2. Normal quantities of III (Methods) added to partly purified component fractions had no significant effect on anaerobic gel band patterns. In some blank gels loaded with excess III, postelectrophoretic destaining with 7% acetic acid left a very faint residual band at the tracking dye position; this was not observed with methanol-acetic acid destaining solutions and normal dye loads. CONCLUSION

The anaerobically modified electrophoresis cell described here has been successfully evaluated in analyses of A. vinelandii nitrogenase and should prove useful with other O,-sensitive proteins, or, more generally, whenever it is desired to perform gel electrophoresis in a specific atmosphere. Guaiazulene-I-sulfonate, III, appears to be an effective and convenient marker dye for following the progress of electrophoretic separations in anaerobic gels treated with a strong reductant, such as dithionite. ACKNOWLEDGMENTS This work was supported by grants from the Research Corporation and by an NIH BSSG Award (No. RR07012-09). We thank Dr. V. K. Shah for communicating details of gel system C.

REFERENCES 1. For examples, see (a) Chrambach, A., and Rodbard, D. (1971) Science 172, 440-451: (b) Hedrick, J. L., and Smith, A. J. (1968) Arch. Eiochem. Biophys. 126, 155-164; (c) Weber, K., and Osborn, M. (1969)5. Biol. Chem. 244,4406-4412; (d) Rodbard, D., and Chrambach, A. (1971) Anal. Biochem. 40, 95-134. 2. McKenna, C. E., McKenna, M-C., and Higa, M. T. (1976) J. Amer. Chem. Sot. 98, 4657-4659. 3.

Fieser, L. (1964) Organic Experiments,

p. 29, D. C. Heath, Boston.

ANAEROBIC

GEL ELECTROPHORESlS

345

4. Meier, W., Meuche, D., and Heilbronner, E. (1963) Helv. Chim. Acta 46, 1929-1940. 5. Davis, B. J. (1964) Ann. N. Y. Acnd. Sci. 121, 404-427. 6. Davis, L. C., Shah, V. K., Brill, W. J., and Orme-Johnson, W. H. (1972) Biochim. Biophys. Acta. 256, 512-523. 7. For gel electrofocusing of hemoglobins in the presence of dithionite, see Park, C. M. (1973) Ann. N. Y. Acad. Sci. 209, 237-257. 8. Dirksen, M. L., and Chrambach, A. (1972) &par. Sci. 7, 747-772. 9. Starchenkov, E. P., and Zhelyuk, V. M. (1976) Prikl. Biokhim. Mikrobiol. 12, 795-797. 10. Chopard-dit-Jean, L. H., and Heilbronner, E. (1953) Helv. Chim. Acfa 36, 144-160. I I. Bernal, I., Rieger, P. H., and Fraenkel, G. K. (1962) J. Chem. Phys. 37, 1489-1495. 12. Shah, V. K., and Brill, W. J. (1973) Biochim. Biophys. Acta 305, 445-454. 13. Eady, R. R., Smith, B. E., Cook, K. A., and Postgate, J. R. (1972) Biochem. J. 128, 655-67s. 14. Israel, D. W., Howard, R. L., Evans, H. J., and Russell, S. A. (1974)J. Bid/. Chem. 249, 500-508.