Comp. Biochem, Phl~iol,. Vol. 62B, pp. 393 lo 402.
0305_0491 79 (k101.0393S02.00~0
© Pert~amon Press Lid 1979. Printed in Great Britain
PURIFICATION AND CHARACTERIZATION OF THE M Y O G L O B I N S O F P A R A M E C I U M T E T R A U R E L I A El)WARD STEERS JR and RICHARD H. DAVIS JR Laboratory of Chemical Biology, NIAMDD, 9000 Rockville Pike, Bethesda, MD 20014, U.S.A. (Received 30 June 1978) Abstract--l. The molecular variants of the myoglobins of Paramecium tetraurelia have been purified as five separate components and designated Mb 1 to Mb 5. 2, The five molecular forms are homogeneous on polyacrylamide gel electrophoresis (PAGE) and isoelectric focusing (IF). 3. The myoglobin species appear to have identical molecular weights and amino acid compositions and differ only in their isoelectric points and relative concentrations in t~ivo. 4. The myoglobin species have an apparent molecular weight of 15,000 + 500 and possess a single heme group per mole which appears to be protoporphyrin IX. 5. The amino acid composition of the five species is: lys12, his 3, arg4, asplT, thrg, ser6, glu16, pro,t, glylt, ala15, cys0, vals, met2, ileT, leul0, tyr5, phe7. 6. The spectra of several ferrous and ferric derivatives of the mixture Mb 1 Mb 5 are presented.
INTRODUCTION
The distribution of heme-containing respiratory pigments capable of reversibly binding oxygen is well documented throughout the animal kingdom (Prosser & Brown, 1961). Their occurrence ranges from various mammalian forms to certain species of the Protozoa. In the ciliate protozoan Paramecium, a hemeoprotein was first reported by Sato & Tamiya (1937) and later by Keilin & Ryley (1953) on the basis of spectroscopic observations on whole cell suspensions. Smith et al. (1962) subsequently reported observations on a partially purified solution of the "hemoglobin" from Paramecium tetraurelia. These authors estimated a molecular weight of 13,000 for their sample on the basis of sedimentation and diffusion measurements for the partially purified material. While it has now been clearly demonstrated that Paramecia possess a heme-containing respiratory protein which is capable of reversibly binding oxygen, no systematic study has been performed on the isolation and characterization of this pigment. Davis & Steers (1976) have recently reported the presence of five weUdefined electrophoretic forms of the heme-protein in P. tetraurelia previously referred to as "hemoglobin" and suggested they be called myoglobins. These five forms appeared to have identical molecular weights of 16,000 on sodium dodecylsulfate disc gel electrophoresis, and isoelectric points varying from 4.16 to 5.54. The present study reports further results on the purification and characterization of the previously identified electrophoretic variants of Paramecium myoglobin.
MATERIALS AND METHODS
All cultures were grown from single cell isolations of Paramecium tetraurelia (variety 4), strain 51.7s. Culture fluid consisted of 0.15 or 0.30~ grass infusion (Cerophyl, Cerophyl Laboratories, Inc., Kansas City, MO) buffered with 0.10~o Na2HPO4"12 H20. Cultures were grown in
5-gallon pyrex carboys containing 101. of culture fluid which had been inoculated 20 hr prior to using with Klebsiella cloacae, strain P. Cultures were expanded by feeding for one doubling every other day with 0.30To grass infusion until a total of eight carboys (801.) was obtained. This procedure yielded cultures containing an average of 3000 cells/ml and a total of 20°24 ml of packed cells upon harvesting. Cells were harvested by passing the culture fluid through a commercial DeLaval cream separator which had been modified according to the procedure of Preer & Preer (1959). This concentrated cell suspension (approx 1600 ml) was further concentrated by centrifugation at 1200g for 2 min in an oil testing centrifuge. The packed cells were resuspended in 3X volume of 0.2 M ratiinose, 0.05 M phosphate buffer, pH 7.2, 0.1 M NaC1 at 0-4°C. The raflinose suspended cells (86 ml) were homogenized by hand using a Dounce-type homogenizer. The resulting brei was examined microscopically to insure that maximum rupture of the cell suspension had occurred. The homogenate was then centrifuged at 46,000g in a Sorvall RC-2B centrifuge for 30min, after which the supernatant solution was adjusted to 50~o saturation with solid ammonium sulfate. The salt solution was centrifuged again at 46,000g for 5 min and the resulting supernatant adjusted to a final concentration of 65~ ammonium sulfate. Following centrifugation of the 50065To ammonium sulfate fraction, the pellet was dissolved in 0.05 M phosphate buffer, pH 7.2 containing 0.02~o K3FE(CN)6; 0.02To KCN (Van Assendelft, 1970). This converted the pigment to the more stable ferric cyanometmyoglobin form. For spectrophotometric studies, aliquots were kept in buffer which did not contain KaFE(CN)6. The fraction was then applied to a Sephadex G-75 column (2 x 75 cm) equilibrated with 0.05 M phosphate buffer, pH 7.2 containing 0.02~ KCN. The effluent was collected in 0.8 ml fractions and read spectrophotometrically at 280 and 416 nm. The appropriate fractions were pooled and precipitated by the addition of solid ammonium sulfate to a final concentration of 65To. The fraction, designated G-75 peak III, was then centrifuged at 46,000 g 5 min in a Sorvall RC-2B and the pellet redissolved in approx 1 ml of the column buffer. The sample was applied to a Sephadex G-50 column (2 x 75 cm) equilibrated with the same buffer used for the G-75 Sephadex column and eluted in 0.8 ml fractions. The effluent was measured spectrophotometrically at 280 and 416nm and
393
394
El)WARD STliERS JR and RICHARD H. DAVIS JR
the appropriate fractions pooled and precipitated with solid ammonium sulfate at a final salt concentration of 65%. At this stage of the purification the myoglobin was judged to be approx 65% pure. Electrophoresis in standard 7.5% polyacrylamide disc gels was carried out according to the procedure described in the manual supplied by the Canal Industrial Corp. (Bethesda, MD). Samples of 100 2 containing from 150 to 250pg of myoglobin were electrophoresed. Following electrophoresis, the pigmented bands (Mb l~Mb 51 were cut from several gels run in parallel and eluted in 0.05 M phosphate buffer at 0-4°C. The elution buffer, 0.5 ml aliquots, was changed 3 times and the final volume (1.5 ml) pooled and dialyzed for 24-40 hr against 0.05 M phosphate buffer to remove gel reagents. Aliquots of the individual myoglobin components eluted in this manner were re-electrophoresed to test for purity. Sucrose gradient isoelectric focusing was performed using pH 4-6 carrier ampholyte (LKB) by a modification of the procedure outlined in the LKB 8100 Ampholine Instruction Manual (LKB-Producter AB, Sweden). A sucrose gradient (50%) was layered in a column (1.5 × 18 cm) to a total volume of 30ml. The column was sealed by polymerizing a 2 cm plug of 5?/0 polyacrylamide gel at the base. The sample, containing 5 8 mg of myoglobin, was added to the light solution (distilled water) prior to forming the density gradient. The column was then fitted to a standard disc electrophoresis apparatus which had been modified to accept the large diameter column. Electrophoresis was carried out at 2-6°C with a constant voltage of 250 V until the current dropped to less than I M A - usually 5-6 hr. Samples were collected by inserting a 20 gauge hypodermic needle into a side-arm attached 2 cm from the base of the column and collecting 12-14 drops per sample or approx 0.5 ml. The molecular weight determinations were carried out using a Spinco Model E Analytical Ultracentrifuge equipped with an electronic speed control and RTIC temperature control unit. Molecular weights were calculated from plots of log dy (y is the fringe displacement in /zm) vs x z (x is the distance in centimeters from rotor center) using the method of Yphantis (1964), where M-
2RT × 2.303 (1 -/~plW 2
×
logdy dx 2 "
The equilibrium patterns were photographed using Kodak metallographic plates and measured with a Nikon Model C comparator. Molecular weights were determined for a
mixture of Mb l - M b 5 as well as for individual myoglobin fractions purified from polyacrylamide disc gels. A speed range of 40,000--48,000 rev/min was used. The samples were dialyzed against a 0.05 M phosphate buffer, 0.1 M NaCI, pH 7.3. The individual myoglobin fractions were hydrolyzed in 5.7 N constant boiling HCI in thick-walled glass tubes which had been thoroughly evacuated before sealing. The hydrolyzates were dried under vacuum and analyzed on a Beckman Model 121 MB automatic amino acid analyzer equipped with a Beckman System AA automatic integrator. All samples were 25 hr hydrolyzates and were uncorrected for possible destruction of individual amino acids during hydrolysis. Residues were calculated on a basis of a monomer molecular weight of 15,000. The spectrum of samples including various derivatives was performed with a Cary Model 15 automatic recording spectrophotometer. Column fractions were read spectrophotimetrically at 280 and 416nm using a Zeiss Model PMQ spectrophotometer.
RESULTS
Purification and physical characterization of the myoglobins of Paramecium tetraurelia The results of the purification procedure using a single lot of cells (lot n u m b e r 82 consisting of 25 ml of packed cells) are summarized in Table 1. This procedure results in an overall purification of approx 500-fold over the starting cell homogenate. The purification estimate is based on the assumption that the starting homogenate in step 1 contains the same a m o u n t of myoglobin as measured in the supernatant of step 2. The large dilution factor together with the slight turbidity encountered at this step makes quantitative spectrophotometric estimates somewhat variable and therefore unreliable. Beginning with step 2 these estimates are consistent and considerably more reliable. Following step 3, the myoglobin fraction is converted to the cyanmetmyoglobin form, except for aliquots removed for spectrophotometric analysis, by dissolving the a m m o n i u m sulfate precipitate in buffer containing K3Fe(CN)6 and K C N (Van Assendelft, 1970). The cyanmetmyoglobin form is presumably the
Table 1. Purification of Paramecium myoglobin Step
i.
Cell Homogenate
2.
Supernatent
Protein
(Mg) a
Mb(M~) b
'lh as % of total
Purification (~ fold)
2200
6.0
~.2
1
700
6,0
0.8
4
fractionation 50-65%
65
4.2
6.5
32
4,
Sephadex G-75
i0
4.0
40,0
200
5.
Sephadex G-50
6
3,6
60.0
300
6a.
PAGE c
i
1.0
~10O.O
~500
6b.
Isoelectric Focusing c
5
5.0
%100.0
%500
3.
(NH4)2SO4
a Protein determined by the method of Lowry et al. (1951). b Estimated spectrophotimetrically assuming an extinction coefficient of cyanmetmyoglobin at 540nm equal to ll.0 (Antonini & Brunori, 1971). c Recovery from electrophoresis varies between 90 and 100~o of the total myoglobin material applied.
Myoglobins of Paramecium
395
1.4
2.4
1.2
2.0
/
E c
o
1.6
o 0
1.2
~
I~
o
0.8 o 0.6
0.4
0 0 ....
1.0
i "4
~%,
,IO, ~ i ~ . ~ 30 -,,,..,..,,r,~ . .50. , , I60- - - -70- . - S80~ 20 40
0.2
90
0
'00
TUBE NUMBER
Fig. 1. Chromatography of the ammonium sulfate fraction (50-65% saturation) on a column (2 x 75 cm) of Sephadex G-75. Elution buffer consisted 0.05 M phosphate buffer, pH 7.2 containing 0.02% KCN. Effluent collected in 0.8 ml fractions. H OD2so rim, O - - O OD416 nm. Pooled fractions designated by Roman numerals: I, lI, IlI.
T
a
m
b
c
d
Fig. 2. Polyacrylamide gel electropherograms of pooled Sephadex G-75 fractions I and lII. Gels are the standard 7.5% polyacrylamide stained for total protein with amido-Schwartz (a) and (c), stained for heme with benzidine HCI (Haut et al., 1962).
396
El)WARD STEERS JR a n d RICHARD H. DAVIS JR
the major myogiobin fraction (III-B) with respect to the heme-containing bands. The percentage of contaminating protein is significantly higher in the leading III-A fraction which has a 280:416 ratio of 2.60 vs a ratio of 0.70 for fraction III-B and, therefore, was not included in the III-B fraction for further purification. The material obtained from the G-50 step is approx 60% pure and represents a 300-fold purification over the starting homogenate of step 1. Figure 4 A,B shows the polyacrylamide gel patterns for both unstained gel and gel stained for total protein with amido-Schwartz of the G-50 pooled fraction III-B. As described previously (Davis & Steers, 1976), the myoglobin fraction from Paramecium consists of five electrophoretic species which consist of two major components (Mb 1 and Mb 4) and three minor components (Mb 2, Mb 3 and Mb 5). Densitometric scans of unstained gels and gels stained specifically for heme show that Mb 1 and Mb 4 account for 65% of the total heine-containing protein while Mb 2, Mb 3 and Mb 5 account for the remaining 355'0. While G-50 chromatography results in a 300-fold purification over the starting material, a major contaminant is still evident near the Mb 3 band (Fig. 4 B) requiring an additional step. The final step in purification was accomplished by eluting the respective heme-containing bands from standard 7.5% polyacrylamide disc gels or, alternatively, by isoelectric focusing on a modified column (2.5 x 15 cm) using a pH 4-6 gradient in 50% sucrose. In the first method, the pigmented bands were carefully cut from a series of standard 7.5% disc gels and
more stable derivative of those hemoglobins and myoglobins studied to date. The ammonium sulfate fraction (50-65%) of step 3 is dissolved in 0.05 M phosphate buffer, pH 7.2, containing 0.02% K3FE(CN)6, 0.02% KCN and chromatographed on a column (1.5 x 75 cm) of Sephadex G-75. A typical elution pattern from this column is shown in Fig. 1. The bulk of the applied protein emerges in the void volume or slightly behind the void volume while the myoglobin fraction is retarded. While the elution profile appears to contain a leading myoglobin shoulder (designated II in Fig. 1), fraction II and III give identical patterns of polyacrylamide gels with respect to their myoglobin components. Fraction II contains significantly higher amounts of contaminating protein than fraction III and was therefore not included in the pooled fraction III for further purification. Figure 2 shows disc gel patterns of the pooled fractions I and III from the Sephadex G-75 column which were stained for total protein and for heme, Fraction III at this stage of purification contains approx 15% of the total protein of the applied sample. The myoglobin at step 3 is approx 40% pure. Following G-75 chromatography, fraction III was chromatographed on Sephadex G-50 (1.0 × 75 cm) using the same elution buffer employed for the G-75 fractionation (Fig. 3). Again, a leading shoulder is seen in the elution pattern from the G-50 column, similar to that seen on the G-75 column. Polyacrylamide gel electrophoresis of this fraction (Ill-A) is indistinguishable from the gel pattern obtained from 2.0
2.0
I
-B
I
I
1.6
t.6
1.2
1.2 O
0
I
I
o o
0.8
0.8
I III-A/i
0.4
0.4
0
. . . . . 0
I0
20
'30
I
I
40
50
TUBE
I
I
60
70
_
BO
90
I00
NUMBER
Fig. 3. Chromatography of Sephadex G-75 fraction II1 on a Sephadex G-50 column (2 × 75 cm). Elution buffer consisted of 0.05 M phosphate buffer, pH 7.2 containing 0.02% KCN. Effluent collected in 0.8ml fractions and read spectrophotometrically. Q, OD2ao nm, O, OD4~6 nm. Pooled fractions designated III-A and III-B.
Myoglobins of Paramecium
A
397
13
2 -.D-
5 --D.-
4
Fig. 4. Polyacrylamide gel electropherograms of the pooled Sephadex G-50 fraction III-B. Gels are standard 7.5~o polyacrylamide stained for total protein with amido-Schwartz (B), and unstained (A).
several slices (usually 10--16) were pooled, broken up with a glass stirring rod, and eluted with phosphate buffer overnight at 4°C. Recovery of the myoglobin by this procedure usually exceeds 80~ of the applied material. Figure 5 shows the results of an electrophoresis experiment in which the pigmented bands were eluted from 10 acrylamide gels, pooled and aliquots re-electrophoresed. The upper set of gels shows the unstained patterns of the separated myoglobins while the lower figure shows the same set of gels stained for total protein with amido-Schwartz. It is evident from these patterns that the five electrophoretic forms of Paramecium myoglobin are homogeneous. c.t~.P. 62/4~
G
Alternatively, it was possible to separate the five electrophoretic species by isoelectric focusing of the Sephadex G-50 fraction. The five components, Mb 1Mb 5, separate into the same relative pattern as seen on the standard polyacrylamide gels, reflecting differences of net charge rather than molecular weight. The isoelectric points for the five species isolated by this procedure were determined to be 5.54, 5.18, 4.98, 4.51 and 4.16, respectively (Davis & Steers, 1970). Samples were collected drop-wise in 0.5 ml fractions from the isoelectric focusing column following completion of a run and assayed spectophotometrically at 280 and 416nm. The appropriate fractions
EDWARD STEERS JR and RICHARD H. DAVIS JR
398
BAND NUMBER I
2
3
4
5
a lad Z
Fig. 5. Polyacrylamide gel electropherograms of the final myoglobin fractions Mb 1-Mb 5 purified by eluting from standard 7.5~o PAG. Upper set of gels shows the unstained patterns of Mb 1-Mb 5. Lower set of gels shows the gels stained for total protein with amido-Schwartz.
were pooled and precipitated at 65~o saturation with (NH4)2SO4 and stored until further use as a precipitate. The final step in purification using electrophoresis represents a 500-fold purification over the starting cell homogenate. The exceedingly small amounts of myoglobin relative to the total protein content of the starting material make the eleetrophoresis methods especially desirable in handling the 2-4 mg yields obtained from a single harvest of cells. Spectra Spectra were obtained for several common derivatives of the myoglobin fraction over the wavelength
range of 350-700 nm. The spectral studies were performed on the concentrated fraction obtained from step 3 in the purification [(NH4)2SO4 precipitation] which had not been converted to the cyanmetmyoglobin form, but was dissolved separately in 0.05 M phosphate buffer at pH 7.2. Table 2 lists the wavelengths of the maximum absorption for several derivatives of this sample while Fig. 6(a) and (b) shows spectra for the ferrous and ferric derivatives. These data and those listed in Table 2 are similar to those previously reported by Smith et al. (1962), and comparable to other mammalian hemoglobins and myoglobins. The absorption spectra of the acid and alkaline ferric myoglobin, however,
Myoglobins of Paramecium
399
Table 2 Derivative
~
8
Soret
Ferrous Derivatives Oxymyoglobin
5B0
542
416
Carbonmonoxo~-myoglobln
570
540
421
Myoglobin
572
422
545
418
Ferric Derivatives Cyanide metmyoglobin Acid metmyoglobin
632
Alkaline metmyoglobin
1.4
I
540
407
540
408
I
I
(a)
1.2
0.6
l
1.0
0.5
o
i
~0.8-
0.4 f"
l
_u 0.6I,-
0
t
0.3 ~ .-4
0,,.
o
i-/.-,.i, i "..1" ~:." i....-..¢, ............ \ " i,,. \.~.
0.4-
0.2
i
0.1 ~
I
4O0 1.4
:
~
,
.
500 X (nm)
T
_
.
600
I
I
(b) 1.2
!i
~0.6
f', :
0.5
1.0 >I--
O "--I
0.4 ),
~zO.8
,ii!it ~- 0.6 a.
,..\.
~t".
o
r-
"\ \.
i
0.4
• ",
0.3 ~
/ ',
.--t
,
". •.....~ . •
"~,~/ \ ,,,¢/ i.........,
0.2
0.1
0.2 i 400
I 50O
I 600
i0
X (nm) Fig. 6. Spectral properties of Paramecium tetraurelia myoglobin as ferrous (a) and ferric (b) derivatives. (a) . . . . . , oxymyoglobin; - - - , carbonmonoxymyoglobin; . . . . . , myoglobin; . . . . . , cyanide haemochromogen, Paramecium. (b) - - - , cyanide metrnyoglobin; . . . . . , acid metmyoglobin: . . . . . , alkaline metmyoglobin; - - - , pyridine haemochromogen, Paramecium.
400
EI)WARI)
STEERS
JR
and
RICHARD
Table 3 Pvrtdtne
H.
DAVIS
JR
Table 4. Molecular weight determination by sedimentation equilibrium
haemochromogen Paramecium ferroprotopogphyrtn Cytochrome g
IXa '
555 557 550
525 525 520
420 418 414
Paramecium ferroprotop%phvrin I x a Cytochrome C
565 565 555
532 536 525
431 434 420
C y a n i d e haemochromogen
q4~){)
Brunori (1971). Vernon & Kamen (1954).
J Antonini
]
1',4ol)
-
-
I:,~l~
i
I f, LiiCl
I , ,i}rJ
,',.}
&
A v e r a g e for Individual >lkogI~,h i n~
differ from those of most other acid and alkaline ferric derivatives in the visible region of the spectrum. Acid ferric metmyoglobin from most species has a wellresolved peak around 500nm and a smaller peak around 635 nm. Paramecium acid ferric metmyoglobin differs from this in the absence of a maxima near 500nm, but shows a large well-resolved peak at 540 nm. The alkaline ferric metmyoglobin of Paramecium differs from the acid ferric metmyoglobin of Paramecium in the absence of a maxima at 632 nm. Alkaline ferric metmyoglobin is usually characterized by a peak near 540nm and broad peak in the 575-595 nm region which is absent in the Paramecium alkaline ferric derivative. A second difference is seen in the fl-band which has a higher absorbance than the a-band. An observation similar to this has been reported by Wittenberg et al. (1965) for the hemoglobin isolated from the perienteric fluid of the nematode Ascaris lumbricoides. The findings reported here confirm the earlier data of Smith et al. (1962), using partially purified samples. The nature of the heme prosthetic group appears to be protoprophyrin IX as seen by the absorption spectra listed in Table 3. The alkaline reduced cyanide and pyridine derivatives of Paramecium myoglobin compare closely with the same derivatives of proto-
\verage
for
all
] 46(FI
'oeter=inati,,ns:
15204)
I ",/r){l
]5rlOrl + 5~)r)
. . . . . . . . . . . . . . . . . . . . .
porphyrin IX (Antonini & Brunori, 1971) while significantly different from cytochrome C which may occasionally occur as a contaminant in Paramecium preparations. Sedimentation equilibrium studies
The electrophoretically purified myoglobin was analyzed by equilibrium centrifugation as both a mixture (Mb 1-Mb 5) of the five electrophoretic forms and individually as separate components. Figure 7 shows a log plot of fringe displacement (log c) against the square of the radial distance X2(CM 2) for one of the experimental runs for a mixture of the purified forms Mb l - M b 5. A straight line plot was observed for this sample as was observed for similar analyses for the individual myoglobins indicative of molecular homogeneity. The mixture illustrated in Fig. 7 resulted in a calculated molecular weight of 14,900 + 300. Table 4 lists molecular weights obtained for the five components in 11 separate measurements. The average molecular weight values range from 14,500 to 15,500. These results lead to
I
I
I
I 48.5
49.0
4,9.5 X2(CM 2)
I
I
50.0
I 50.,5
3.2-
3.0
2.8
2.6 U
g .J 2.4
2.2
2.0
1.8
Fig. 7. Molecular weight determination of the purified Mb l-Mb 5 mixture by the high speed equilibrium technique of Yphantis (1964). Protein concentration in the above run was 0.3 mg/ml in 0.05 M phosphate buffer, pH 7.2, 0.IOM NaCI. Speed for the run was 48,000rev/min. A molecular weight of 14,900 + 300 was calculated from the above plot.
401
Myoglobins of Paramecium Table 5. Comparison of the amino acid compositions of Paramecium myoglobins* Mb-I
Mb-2
Mb-3
Mb-4
Mb-5
Mbl-Mb5
Amino Acid
lys
12
12
12
12
12
his
4
3
3
3
3
3
arg
3
4
4
3
4
4
asp
18
17
17
19
16
17
thr
9
7
8
Ii
8
9
ser
5
5
6
4
5
6
glu
16
17
17
15
15
16
pro
4
6
5
3
6
4
gly
11
i0
ii
12
ii
ii
ala
14
12
13
15
15
15
cys
0
0
0
O
0
0
val
8
8
9
8
9
8
met
2
2
2
i
2
2
lle
7
8
8
7
8
7
leu
i0
ii
ii
9
9
i0
tyr
5
5
5
6
5
5
phe
8
7
7
i0
7
7
136
134
138
138
135
136
15039
14940
15258
15251
14694
14950
Number of amino acids Molecular weight
12
* Calculated on the basis of 12 lysine residues per mole Number of residues rounded to the nearest integer. the conclusion that the myoglobin forms have identical molecular weights of 15,000 + 500, which is in close agreement with the previously reported results of 16,000 (Davis & Steers, 1970) obtained from Na dodecyl sulfate gel electrophoresis, where it was concluded a single molecular weight species existed for the five forms. Chemical properties
The electrophoretically purified myoglobins were analyzed for their amino acid composition following hydrolysis for 25 hr in 5.7 N HC1. Table 5 lists the amino acid composition of the combined species as well as the individual forms. The residues were calculated on the basis of 12 lysines/mole of molecular weight 15,000. On this basis, the five proteins contain from 135 to 138 amino acid residues with molecular weights ranging from 14,694 to 15,258. The five species are further characterized by the absence of cysteine, a finding which is characteristic of the mammalian myoglobins (Dayoff, 1976), but not of many of the invertebrate myoglobins (Terwillinger & Read, 1969, 1970; Read, 1968). The amino acid compositions of the five species were compared for compositional relatedness to each other by the method of Harris & Teller (1973) using the formula:
No significant differences between the molecular forms could be found by this method. In this method, a composition divergence (D) of less than 7.0 x 102 is indicative of a "non-fortuitous similarity of compo-
sition" according to Harris and Teller. Values of D for the five species as well as the mixture Mb l - M b 5 when compared to each other ranged from 1.5 to 4.7 x 10 2, with an average value of 3.0 × 10 2. While these values are significantly lower than the maximum value for relatedness determined by Harris and Teller, conclusions concerning sequential identity must await such analysis. Clearly, the amino acid compositions do not show any significant differences at the level of analyses reported here. DISCUSSION
The partial purification of Paramecium myoglobin was reported by Smith et aL (1962), but full purity of the protein(s) was clearly not obtained. In a still later study, Davis & Steers (1976) reported the partial purification of Paramecium myoglobin and its separation into five distinct electrophoretic species. The present study describes the full purification together with several of the physical and chemical properties for the five electrophoretic forms of Paramecium myoglobin. The method described in this study utilizing electrophoretic techniques results in an overall purification of approx 500-fold. Large-scale purification of Paramecium myoglobin is limited by the in vivo levels of the protein and to the maximum volume of culture that can be successfully handled in growing the cells. In the present study, 25 cm 3 of packed cells harvested from 801. of culture fluid yielded an estimated total of 3.6 mg of myoglobin or approx 45/Lg of total myoglobin per liter of culture. A great variability in the myoglobin concentration from culture to culture has been observed, for which no basis has been obvious to date. The total myoglobin yield from harvest to harvest seldom exceeds 4mg/801. culture and drops as low as 1 mg/oulture lot. The conditions employed for growing Paramecium limit the maximum volume of cultures to between 160 and 2001. which can be expected to yield myoglobin levels of approx 2-8 mg/ harvest. Such low levels of material, when taken together with the close isoelectric points of the five molecular forms, prompted our use of electrophoresis as a final step in purification. The visual identification of the pigmented bands in either of the electrophoretic techniques employed here with their near equidistant spacing resulted in the relative ease of eluting highly purified fractions. The molecular weight determinations on the isolated species showed no significant differences. Moreover, mixtures of the various electrophoretic forms gave straight-line plots for log C vs XZ(cm2), indicative of homogeneity of the preparations under conditions similar to those used in electrophoresis. The average molecular weight value calculated for all determinations by sedimentation equilibrium was 15,000 + 500, which agrees closely with the previously determined value of 16,000 obtained by SDS-gel electrophoresis (Davis & Steers, 1976). The amino acid composition does not appear to reflect any significant differences for the 17 residues analyzed. The spectral properties are very similar to those of hemoglobins and myoglobins characterized to date which lead to the conclusion that the five electro-
402
El)WARD STEERSJR and RI('HARD H. DAVISJR
phoretic species isolated from P. tetraurelia are monoheme, single polypeptide chains of approx 15,000 molecular weight. The only definitive differences observed to date among the five forms is in the relative amount in which they are isolated and in their isoelectric points. The values, previously determined by us (Davis & o/ respectively, Steers, 1976), are 32, 11, 12, 33 and 12/,,, for Mb l - M b 5 with isoelectric points of 5.54, 5.18, 4.98, 4.51 and 4.16. The quantitative differences between the five species (Mb 1 Mb 5) has proven to be consistent for several isolations suggesting an in vivo difference. The amino acid composition of the five proteins was analyzed for compositional relatedness by the method of Harris & Teller (1973) with no significant differences evident between the five species. The molecular weight studies and amino acid analyses suggest that the myoglobin species in Paramecium represent only electrophoretic variants. However, this method of analysis is not sensitive enough to detect small differences in primary sequence which may exist and thereby escape detection by the techniques used in this study. In addition, the possible biological significance of variant forms of the myoglobin is not evident. The exact function of these heme-proteins is still unknown. Smith et al. (1962) concluded that "long term" storage function was not likely based on calculations of oxygen consumption and cell numbers. By definition, the heine-proteins of P. tetraurelia are intracellular and therefore not "circulatory" in the usual sense. Earlier suggestions (Smith et al. 1962) that myoglobins (or hemoglobins) with high oxygen affinity may enhance the efficiency of the cytochrome system under conditions of reduced oxygen seem most probable at present. The significance of several electrophoretic species with presumably the same function is even less well understood. Such questions of the possible biological function and significance of the multiple forms of myoglobin in Paramecium await further studies which are now possible with a procedure to isolate the forms in a highly purified state. Acknowledgement--The authors would like to acknowledge the special technical assistance of Mr Clifford E. Lee
in the isolation, cultivation and harvesting of cell cultures and various other aspects of this work. REFERENCES
ANTON1NI E. & BRUNORI M. (1971) Hemoglobin and Myoglobin in their Reactions witl7 Ligands. Frontiers of Biology, Vol. 21 Elsevier, Amsterdam. DAvis R. H. and STEERS E. JR (1976) Myoglobin from the ciliate protozoan Paramecium aurelia. Comp. Biochem. Physiol. 54B, 141 143. HAUT A., Tel)HOPE G. R., CARTWRIGHTG. E. & WINTROBE M. M. (1962) The nonhemoglobin erythrocytic proteins studied by electrophoresis on starch gel. J. clin. hlvest. 41, 57%587. KE1LIN D. & RYLEYJ. F. (1953) Haemoglobin in protozoa. Nature, LoAd. 172, 45l. LOWRY O, H., ROSEBROUGHN. J., FARR A. L. & RANDALL R. J. (1951). Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275. PREER J. R. & PREER L. B. (1959) Gel diffusion studies on the antigens of isolated cellular components of Parameciunl. J. Protozool. 6, 88 100. PROSSI-R C. L. and BROWN F. A., JR (1961) Comparative Animal Physiology. W. B. Saunders, Philadelphia. READ K. R. H. (1968) The myoglobins of the gastropod molluscs, Busycon contrarium Conrad, Lunatia heros, Say, Littorina littorea, L. and Siphonaria gigas, Sowerby. Comp. Biochem. Biophys. 25, 81-94. SATOT. & TAMIYAH. (19371 Cytologia, Fujii Jubilaei Voluman, pp. 1133 1137. SMITH M. H., GEORGE P. & PREliR J. R., JR (1962) Preliminary observations on isolated Paramecium hemoglobin. Arehs Biochem. Biophys. 99, 313-318. TERWlLLIGER R. C. & REAl) K. R. H. (1969) Quaternary structure of the radular muscle myoglobin of the gastropod mollusc Buccinium undatum L. Comp. Biochem. Physiol. 31, 55 64. TERWJLLIGER R. C. & READ K. R. H. (1970) The radular muscle of the amphineuran molluscs, Katharina tunicata, Wood, Cryptochiton stelleri, Middendorf, and Mopalia muscosa, Gould. Int. J. Biochem. I, 281-291. VAN ASSENDELETO. W. (1970) Spectrophotometry of Hemoglobin Derivatives. Thomas, Netherlands. VERNON L. P. & KAMENM. D. (1954) Hematin compounds in photosynthetic bacteria. J. biol. Chem. 211, 643-662. WITTENBERGB. A., OKAZAKIT. & WITTENBERGJ. B. (1965) The hemoglobin of ascaris perienteric fluid. I. Purification and spectra. Biochem. biophys. Acta III, 485-495. YPHANTIS D. A. (1964) Equilibrium ultracentrifugation of dilute solutions. Biochemistry 3, 297-317.
0305_0491 79 (k101.0393S02.00~0
© Pert~amon Press Lid 1979. Printed in Great Britain
PURIFICATION AND CHARACTERIZATION OF THE M Y O G L O B I N S O F P A R A M E C I U M T E T R A U R E L I A El)WARD STEERS JR and RICHARD H. DAVIS JR Laboratory of Chemical Biology, NIAMDD, 9000 Rockville Pike, Bethesda, MD 20014, U.S.A. (Received 30 June 1978) Abstract--l. The molecular variants of the myoglobins of Paramecium tetraurelia have been purified as five separate components and designated Mb 1 to Mb 5. 2, The five molecular forms are homogeneous on polyacrylamide gel electrophoresis (PAGE) and isoelectric focusing (IF). 3. The myoglobin species appear to have identical molecular weights and amino acid compositions and differ only in their isoelectric points and relative concentrations in t~ivo. 4. The myoglobin species have an apparent molecular weight of 15,000 + 500 and possess a single heme group per mole which appears to be protoporphyrin IX. 5. The amino acid composition of the five species is: lys12, his 3, arg4, asplT, thrg, ser6, glu16, pro,t, glylt, ala15, cys0, vals, met2, ileT, leul0, tyr5, phe7. 6. The spectra of several ferrous and ferric derivatives of the mixture Mb 1 Mb 5 are presented.
INTRODUCTION
The distribution of heme-containing respiratory pigments capable of reversibly binding oxygen is well documented throughout the animal kingdom (Prosser & Brown, 1961). Their occurrence ranges from various mammalian forms to certain species of the Protozoa. In the ciliate protozoan Paramecium, a hemeoprotein was first reported by Sato & Tamiya (1937) and later by Keilin & Ryley (1953) on the basis of spectroscopic observations on whole cell suspensions. Smith et al. (1962) subsequently reported observations on a partially purified solution of the "hemoglobin" from Paramecium tetraurelia. These authors estimated a molecular weight of 13,000 for their sample on the basis of sedimentation and diffusion measurements for the partially purified material. While it has now been clearly demonstrated that Paramecia possess a heme-containing respiratory protein which is capable of reversibly binding oxygen, no systematic study has been performed on the isolation and characterization of this pigment. Davis & Steers (1976) have recently reported the presence of five weUdefined electrophoretic forms of the heme-protein in P. tetraurelia previously referred to as "hemoglobin" and suggested they be called myoglobins. These five forms appeared to have identical molecular weights of 16,000 on sodium dodecylsulfate disc gel electrophoresis, and isoelectric points varying from 4.16 to 5.54. The present study reports further results on the purification and characterization of the previously identified electrophoretic variants of Paramecium myoglobin.
MATERIALS AND METHODS
All cultures were grown from single cell isolations of Paramecium tetraurelia (variety 4), strain 51.7s. Culture fluid consisted of 0.15 or 0.30~ grass infusion (Cerophyl, Cerophyl Laboratories, Inc., Kansas City, MO) buffered with 0.10~o Na2HPO4"12 H20. Cultures were grown in
5-gallon pyrex carboys containing 101. of culture fluid which had been inoculated 20 hr prior to using with Klebsiella cloacae, strain P. Cultures were expanded by feeding for one doubling every other day with 0.30To grass infusion until a total of eight carboys (801.) was obtained. This procedure yielded cultures containing an average of 3000 cells/ml and a total of 20°24 ml of packed cells upon harvesting. Cells were harvested by passing the culture fluid through a commercial DeLaval cream separator which had been modified according to the procedure of Preer & Preer (1959). This concentrated cell suspension (approx 1600 ml) was further concentrated by centrifugation at 1200g for 2 min in an oil testing centrifuge. The packed cells were resuspended in 3X volume of 0.2 M ratiinose, 0.05 M phosphate buffer, pH 7.2, 0.1 M NaC1 at 0-4°C. The raflinose suspended cells (86 ml) were homogenized by hand using a Dounce-type homogenizer. The resulting brei was examined microscopically to insure that maximum rupture of the cell suspension had occurred. The homogenate was then centrifuged at 46,000g in a Sorvall RC-2B centrifuge for 30min, after which the supernatant solution was adjusted to 50~o saturation with solid ammonium sulfate. The salt solution was centrifuged again at 46,000g for 5 min and the resulting supernatant adjusted to a final concentration of 65~ ammonium sulfate. Following centrifugation of the 50065To ammonium sulfate fraction, the pellet was dissolved in 0.05 M phosphate buffer, pH 7.2 containing 0.02~o K3FE(CN)6; 0.02To KCN (Van Assendelft, 1970). This converted the pigment to the more stable ferric cyanometmyoglobin form. For spectrophotometric studies, aliquots were kept in buffer which did not contain KaFE(CN)6. The fraction was then applied to a Sephadex G-75 column (2 x 75 cm) equilibrated with 0.05 M phosphate buffer, pH 7.2 containing 0.02~ KCN. The effluent was collected in 0.8 ml fractions and read spectrophotometrically at 280 and 416 nm. The appropriate fractions were pooled and precipitated by the addition of solid ammonium sulfate to a final concentration of 65To. The fraction, designated G-75 peak III, was then centrifuged at 46,000 g 5 min in a Sorvall RC-2B and the pellet redissolved in approx 1 ml of the column buffer. The sample was applied to a Sephadex G-50 column (2 x 75 cm) equilibrated with the same buffer used for the G-75 Sephadex column and eluted in 0.8 ml fractions. The effluent was measured spectrophotometrically at 280 and 416nm and
393
394
El)WARD STliERS JR and RICHARD H. DAVIS JR
the appropriate fractions pooled and precipitated with solid ammonium sulfate at a final salt concentration of 65%. At this stage of the purification the myoglobin was judged to be approx 65% pure. Electrophoresis in standard 7.5% polyacrylamide disc gels was carried out according to the procedure described in the manual supplied by the Canal Industrial Corp. (Bethesda, MD). Samples of 100 2 containing from 150 to 250pg of myoglobin were electrophoresed. Following electrophoresis, the pigmented bands (Mb l~Mb 51 were cut from several gels run in parallel and eluted in 0.05 M phosphate buffer at 0-4°C. The elution buffer, 0.5 ml aliquots, was changed 3 times and the final volume (1.5 ml) pooled and dialyzed for 24-40 hr against 0.05 M phosphate buffer to remove gel reagents. Aliquots of the individual myoglobin components eluted in this manner were re-electrophoresed to test for purity. Sucrose gradient isoelectric focusing was performed using pH 4-6 carrier ampholyte (LKB) by a modification of the procedure outlined in the LKB 8100 Ampholine Instruction Manual (LKB-Producter AB, Sweden). A sucrose gradient (50%) was layered in a column (1.5 × 18 cm) to a total volume of 30ml. The column was sealed by polymerizing a 2 cm plug of 5?/0 polyacrylamide gel at the base. The sample, containing 5 8 mg of myoglobin, was added to the light solution (distilled water) prior to forming the density gradient. The column was then fitted to a standard disc electrophoresis apparatus which had been modified to accept the large diameter column. Electrophoresis was carried out at 2-6°C with a constant voltage of 250 V until the current dropped to less than I M A - usually 5-6 hr. Samples were collected by inserting a 20 gauge hypodermic needle into a side-arm attached 2 cm from the base of the column and collecting 12-14 drops per sample or approx 0.5 ml. The molecular weight determinations were carried out using a Spinco Model E Analytical Ultracentrifuge equipped with an electronic speed control and RTIC temperature control unit. Molecular weights were calculated from plots of log dy (y is the fringe displacement in /zm) vs x z (x is the distance in centimeters from rotor center) using the method of Yphantis (1964), where M-
2RT × 2.303 (1 -/~plW 2
×
logdy dx 2 "
The equilibrium patterns were photographed using Kodak metallographic plates and measured with a Nikon Model C comparator. Molecular weights were determined for a
mixture of Mb l - M b 5 as well as for individual myoglobin fractions purified from polyacrylamide disc gels. A speed range of 40,000--48,000 rev/min was used. The samples were dialyzed against a 0.05 M phosphate buffer, 0.1 M NaCI, pH 7.3. The individual myoglobin fractions were hydrolyzed in 5.7 N constant boiling HCI in thick-walled glass tubes which had been thoroughly evacuated before sealing. The hydrolyzates were dried under vacuum and analyzed on a Beckman Model 121 MB automatic amino acid analyzer equipped with a Beckman System AA automatic integrator. All samples were 25 hr hydrolyzates and were uncorrected for possible destruction of individual amino acids during hydrolysis. Residues were calculated on a basis of a monomer molecular weight of 15,000. The spectrum of samples including various derivatives was performed with a Cary Model 15 automatic recording spectrophotometer. Column fractions were read spectrophotimetrically at 280 and 416nm using a Zeiss Model PMQ spectrophotometer.
RESULTS
Purification and physical characterization of the myoglobins of Paramecium tetraurelia The results of the purification procedure using a single lot of cells (lot n u m b e r 82 consisting of 25 ml of packed cells) are summarized in Table 1. This procedure results in an overall purification of approx 500-fold over the starting cell homogenate. The purification estimate is based on the assumption that the starting homogenate in step 1 contains the same a m o u n t of myoglobin as measured in the supernatant of step 2. The large dilution factor together with the slight turbidity encountered at this step makes quantitative spectrophotometric estimates somewhat variable and therefore unreliable. Beginning with step 2 these estimates are consistent and considerably more reliable. Following step 3, the myoglobin fraction is converted to the cyanmetmyoglobin form, except for aliquots removed for spectrophotometric analysis, by dissolving the a m m o n i u m sulfate precipitate in buffer containing K3Fe(CN)6 and K C N (Van Assendelft, 1970). The cyanmetmyoglobin form is presumably the
Table 1. Purification of Paramecium myoglobin Step
i.
Cell Homogenate
2.
Supernatent
Protein
(Mg) a
Mb(M~) b
'lh as % of total
Purification (~ fold)
2200
6.0
~.2
1
700
6,0
0.8
4
fractionation 50-65%
65
4.2
6.5
32
4,
Sephadex G-75
i0
4.0
40,0
200
5.
Sephadex G-50
6
3,6
60.0
300
6a.
PAGE c
i
1.0
~10O.O
~500
6b.
Isoelectric Focusing c
5
5.0
%100.0
%500
3.
(NH4)2SO4
a Protein determined by the method of Lowry et al. (1951). b Estimated spectrophotimetrically assuming an extinction coefficient of cyanmetmyoglobin at 540nm equal to ll.0 (Antonini & Brunori, 1971). c Recovery from electrophoresis varies between 90 and 100~o of the total myoglobin material applied.
Myoglobins of Paramecium
395
1.4
2.4
1.2
2.0
/
E c
o
1.6
o 0
1.2
~
I~
o
0.8 o 0.6
0.4
0 0 ....
1.0
i "4
~%,
,IO, ~ i ~ . ~ 30 -,,,..,..,,r,~ . .50. , , I60- - - -70- . - S80~ 20 40
0.2
90
0
'00
TUBE NUMBER
Fig. 1. Chromatography of the ammonium sulfate fraction (50-65% saturation) on a column (2 x 75 cm) of Sephadex G-75. Elution buffer consisted 0.05 M phosphate buffer, pH 7.2 containing 0.02% KCN. Effluent collected in 0.8 ml fractions. H OD2so rim, O - - O OD416 nm. Pooled fractions designated by Roman numerals: I, lI, IlI.
T
a
m
b
c
d
Fig. 2. Polyacrylamide gel electropherograms of pooled Sephadex G-75 fractions I and lII. Gels are the standard 7.5% polyacrylamide stained for total protein with amido-Schwartz (a) and (c), stained for heme with benzidine HCI (Haut et al., 1962).
396
El)WARD STEERS JR a n d RICHARD H. DAVIS JR
the major myogiobin fraction (III-B) with respect to the heme-containing bands. The percentage of contaminating protein is significantly higher in the leading III-A fraction which has a 280:416 ratio of 2.60 vs a ratio of 0.70 for fraction III-B and, therefore, was not included in the III-B fraction for further purification. The material obtained from the G-50 step is approx 60% pure and represents a 300-fold purification over the starting homogenate of step 1. Figure 4 A,B shows the polyacrylamide gel patterns for both unstained gel and gel stained for total protein with amido-Schwartz of the G-50 pooled fraction III-B. As described previously (Davis & Steers, 1976), the myoglobin fraction from Paramecium consists of five electrophoretic species which consist of two major components (Mb 1 and Mb 4) and three minor components (Mb 2, Mb 3 and Mb 5). Densitometric scans of unstained gels and gels stained specifically for heme show that Mb 1 and Mb 4 account for 65% of the total heine-containing protein while Mb 2, Mb 3 and Mb 5 account for the remaining 355'0. While G-50 chromatography results in a 300-fold purification over the starting material, a major contaminant is still evident near the Mb 3 band (Fig. 4 B) requiring an additional step. The final step in purification was accomplished by eluting the respective heme-containing bands from standard 7.5% polyacrylamide disc gels or, alternatively, by isoelectric focusing on a modified column (2.5 x 15 cm) using a pH 4-6 gradient in 50% sucrose. In the first method, the pigmented bands were carefully cut from a series of standard 7.5% disc gels and
more stable derivative of those hemoglobins and myoglobins studied to date. The ammonium sulfate fraction (50-65%) of step 3 is dissolved in 0.05 M phosphate buffer, pH 7.2, containing 0.02% K3FE(CN)6, 0.02% KCN and chromatographed on a column (1.5 x 75 cm) of Sephadex G-75. A typical elution pattern from this column is shown in Fig. 1. The bulk of the applied protein emerges in the void volume or slightly behind the void volume while the myoglobin fraction is retarded. While the elution profile appears to contain a leading myoglobin shoulder (designated II in Fig. 1), fraction II and III give identical patterns of polyacrylamide gels with respect to their myoglobin components. Fraction II contains significantly higher amounts of contaminating protein than fraction III and was therefore not included in the pooled fraction III for further purification. Figure 2 shows disc gel patterns of the pooled fractions I and III from the Sephadex G-75 column which were stained for total protein and for heme, Fraction III at this stage of purification contains approx 15% of the total protein of the applied sample. The myoglobin at step 3 is approx 40% pure. Following G-75 chromatography, fraction III was chromatographed on Sephadex G-50 (1.0 × 75 cm) using the same elution buffer employed for the G-75 fractionation (Fig. 3). Again, a leading shoulder is seen in the elution pattern from the G-50 column, similar to that seen on the G-75 column. Polyacrylamide gel electrophoresis of this fraction (Ill-A) is indistinguishable from the gel pattern obtained from 2.0
2.0
I
-B
I
I
1.6
t.6
1.2
1.2 O
0
I
I
o o
0.8
0.8
I III-A/i
0.4
0.4
0
. . . . . 0
I0
20
'30
I
I
40
50
TUBE
I
I
60
70
_
BO
90
I00
NUMBER
Fig. 3. Chromatography of Sephadex G-75 fraction II1 on a Sephadex G-50 column (2 × 75 cm). Elution buffer consisted of 0.05 M phosphate buffer, pH 7.2 containing 0.02% KCN. Effluent collected in 0.8ml fractions and read spectrophotometrically. Q, OD2ao nm, O, OD4~6 nm. Pooled fractions designated III-A and III-B.
Myoglobins of Paramecium
A
397
13
2 -.D-
5 --D.-
4
Fig. 4. Polyacrylamide gel electropherograms of the pooled Sephadex G-50 fraction III-B. Gels are standard 7.5~o polyacrylamide stained for total protein with amido-Schwartz (B), and unstained (A).
several slices (usually 10--16) were pooled, broken up with a glass stirring rod, and eluted with phosphate buffer overnight at 4°C. Recovery of the myoglobin by this procedure usually exceeds 80~ of the applied material. Figure 5 shows the results of an electrophoresis experiment in which the pigmented bands were eluted from 10 acrylamide gels, pooled and aliquots re-electrophoresed. The upper set of gels shows the unstained patterns of the separated myoglobins while the lower figure shows the same set of gels stained for total protein with amido-Schwartz. It is evident from these patterns that the five electrophoretic forms of Paramecium myoglobin are homogeneous. c.t~.P. 62/4~
G
Alternatively, it was possible to separate the five electrophoretic species by isoelectric focusing of the Sephadex G-50 fraction. The five components, Mb 1Mb 5, separate into the same relative pattern as seen on the standard polyacrylamide gels, reflecting differences of net charge rather than molecular weight. The isoelectric points for the five species isolated by this procedure were determined to be 5.54, 5.18, 4.98, 4.51 and 4.16, respectively (Davis & Steers, 1970). Samples were collected drop-wise in 0.5 ml fractions from the isoelectric focusing column following completion of a run and assayed spectophotometrically at 280 and 416nm. The appropriate fractions
EDWARD STEERS JR and RICHARD H. DAVIS JR
398
BAND NUMBER I
2
3
4
5
a lad Z
Fig. 5. Polyacrylamide gel electropherograms of the final myoglobin fractions Mb 1-Mb 5 purified by eluting from standard 7.5~o PAG. Upper set of gels shows the unstained patterns of Mb 1-Mb 5. Lower set of gels shows the gels stained for total protein with amido-Schwartz.
were pooled and precipitated at 65~o saturation with (NH4)2SO4 and stored until further use as a precipitate. The final step in purification using electrophoresis represents a 500-fold purification over the starting cell homogenate. The exceedingly small amounts of myoglobin relative to the total protein content of the starting material make the eleetrophoresis methods especially desirable in handling the 2-4 mg yields obtained from a single harvest of cells. Spectra Spectra were obtained for several common derivatives of the myoglobin fraction over the wavelength
range of 350-700 nm. The spectral studies were performed on the concentrated fraction obtained from step 3 in the purification [(NH4)2SO4 precipitation] which had not been converted to the cyanmetmyoglobin form, but was dissolved separately in 0.05 M phosphate buffer at pH 7.2. Table 2 lists the wavelengths of the maximum absorption for several derivatives of this sample while Fig. 6(a) and (b) shows spectra for the ferrous and ferric derivatives. These data and those listed in Table 2 are similar to those previously reported by Smith et al. (1962), and comparable to other mammalian hemoglobins and myoglobins. The absorption spectra of the acid and alkaline ferric myoglobin, however,
Myoglobins of Paramecium
399
Table 2 Derivative
~
8
Soret
Ferrous Derivatives Oxymyoglobin
5B0
542
416
Carbonmonoxo~-myoglobln
570
540
421
Myoglobin
572
422
545
418
Ferric Derivatives Cyanide metmyoglobin Acid metmyoglobin
632
Alkaline metmyoglobin
1.4
I
540
407
540
408
I
I
(a)
1.2
0.6
l
1.0
0.5
o
i
~0.8-
0.4 f"
l
_u 0.6I,-
0
t
0.3 ~ .-4
0,,.
o
i-/.-,.i, i "..1" ~:." i....-..¢, ............ \ " i,,. \.~.
0.4-
0.2
i
0.1 ~
I
4O0 1.4
:
~
,
.
500 X (nm)
T
_
.
600
I
I
(b) 1.2
!i
~0.6
f', :
0.5
1.0 >I--
O "--I
0.4 ),
~zO.8
,ii!it ~- 0.6 a.
,..\.
~t".
o
r-
"\ \.
i
0.4
• ",
0.3 ~
/ ',
.--t
,
". •.....~ . •
"~,~/ \ ,,,¢/ i.........,
0.2
0.1
0.2 i 400
I 50O
I 600
i0
X (nm) Fig. 6. Spectral properties of Paramecium tetraurelia myoglobin as ferrous (a) and ferric (b) derivatives. (a) . . . . . , oxymyoglobin; - - - , carbonmonoxymyoglobin; . . . . . , myoglobin; . . . . . , cyanide haemochromogen, Paramecium. (b) - - - , cyanide metrnyoglobin; . . . . . , acid metmyoglobin: . . . . . , alkaline metmyoglobin; - - - , pyridine haemochromogen, Paramecium.
400
EI)WARI)
STEERS
JR
and
RICHARD
Table 3 Pvrtdtne
H.
DAVIS
JR
Table 4. Molecular weight determination by sedimentation equilibrium
haemochromogen Paramecium ferroprotopogphyrtn Cytochrome g
IXa '
555 557 550
525 525 520
420 418 414
Paramecium ferroprotop%phvrin I x a Cytochrome C
565 565 555
532 536 525
431 434 420
C y a n i d e haemochromogen
q4~){)
Brunori (1971). Vernon & Kamen (1954).
J Antonini
]
1',4ol)
-
-
I:,~l~
i
I f, LiiCl
I , ,i}rJ
,',.}
&
A v e r a g e for Individual >lkogI~,h i n~
differ from those of most other acid and alkaline ferric derivatives in the visible region of the spectrum. Acid ferric metmyoglobin from most species has a wellresolved peak around 500nm and a smaller peak around 635 nm. Paramecium acid ferric metmyoglobin differs from this in the absence of a maxima near 500nm, but shows a large well-resolved peak at 540 nm. The alkaline ferric metmyoglobin of Paramecium differs from the acid ferric metmyoglobin of Paramecium in the absence of a maxima at 632 nm. Alkaline ferric metmyoglobin is usually characterized by a peak near 540nm and broad peak in the 575-595 nm region which is absent in the Paramecium alkaline ferric derivative. A second difference is seen in the fl-band which has a higher absorbance than the a-band. An observation similar to this has been reported by Wittenberg et al. (1965) for the hemoglobin isolated from the perienteric fluid of the nematode Ascaris lumbricoides. The findings reported here confirm the earlier data of Smith et al. (1962), using partially purified samples. The nature of the heme prosthetic group appears to be protoprophyrin IX as seen by the absorption spectra listed in Table 3. The alkaline reduced cyanide and pyridine derivatives of Paramecium myoglobin compare closely with the same derivatives of proto-
\verage
for
all
] 46(FI
'oeter=inati,,ns:
15204)
I ",/r){l
]5rlOrl + 5~)r)
. . . . . . . . . . . . . . . . . . . . .
porphyrin IX (Antonini & Brunori, 1971) while significantly different from cytochrome C which may occasionally occur as a contaminant in Paramecium preparations. Sedimentation equilibrium studies
The electrophoretically purified myoglobin was analyzed by equilibrium centrifugation as both a mixture (Mb 1-Mb 5) of the five electrophoretic forms and individually as separate components. Figure 7 shows a log plot of fringe displacement (log c) against the square of the radial distance X2(CM 2) for one of the experimental runs for a mixture of the purified forms Mb l - M b 5. A straight line plot was observed for this sample as was observed for similar analyses for the individual myoglobins indicative of molecular homogeneity. The mixture illustrated in Fig. 7 resulted in a calculated molecular weight of 14,900 + 300. Table 4 lists molecular weights obtained for the five components in 11 separate measurements. The average molecular weight values range from 14,500 to 15,500. These results lead to
I
I
I
I 48.5
49.0
4,9.5 X2(CM 2)
I
I
50.0
I 50.,5
3.2-
3.0
2.8
2.6 U
g .J 2.4
2.2
2.0
1.8
Fig. 7. Molecular weight determination of the purified Mb l-Mb 5 mixture by the high speed equilibrium technique of Yphantis (1964). Protein concentration in the above run was 0.3 mg/ml in 0.05 M phosphate buffer, pH 7.2, 0.IOM NaCI. Speed for the run was 48,000rev/min. A molecular weight of 14,900 + 300 was calculated from the above plot.
401
Myoglobins of Paramecium Table 5. Comparison of the amino acid compositions of Paramecium myoglobins* Mb-I
Mb-2
Mb-3
Mb-4
Mb-5
Mbl-Mb5
Amino Acid
lys
12
12
12
12
12
his
4
3
3
3
3
3
arg
3
4
4
3
4
4
asp
18
17
17
19
16
17
thr
9
7
8
Ii
8
9
ser
5
5
6
4
5
6
glu
16
17
17
15
15
16
pro
4
6
5
3
6
4
gly
11
i0
ii
12
ii
ii
ala
14
12
13
15
15
15
cys
0
0
0
O
0
0
val
8
8
9
8
9
8
met
2
2
2
i
2
2
lle
7
8
8
7
8
7
leu
i0
ii
ii
9
9
i0
tyr
5
5
5
6
5
5
phe
8
7
7
i0
7
7
136
134
138
138
135
136
15039
14940
15258
15251
14694
14950
Number of amino acids Molecular weight
12
* Calculated on the basis of 12 lysine residues per mole Number of residues rounded to the nearest integer. the conclusion that the myoglobin forms have identical molecular weights of 15,000 + 500, which is in close agreement with the previously reported results of 16,000 (Davis & Steers, 1970) obtained from Na dodecyl sulfate gel electrophoresis, where it was concluded a single molecular weight species existed for the five forms. Chemical properties
The electrophoretically purified myoglobins were analyzed for their amino acid composition following hydrolysis for 25 hr in 5.7 N HC1. Table 5 lists the amino acid composition of the combined species as well as the individual forms. The residues were calculated on the basis of 12 lysines/mole of molecular weight 15,000. On this basis, the five proteins contain from 135 to 138 amino acid residues with molecular weights ranging from 14,694 to 15,258. The five species are further characterized by the absence of cysteine, a finding which is characteristic of the mammalian myoglobins (Dayoff, 1976), but not of many of the invertebrate myoglobins (Terwillinger & Read, 1969, 1970; Read, 1968). The amino acid compositions of the five species were compared for compositional relatedness to each other by the method of Harris & Teller (1973) using the formula:
No significant differences between the molecular forms could be found by this method. In this method, a composition divergence (D) of less than 7.0 x 102 is indicative of a "non-fortuitous similarity of compo-
sition" according to Harris and Teller. Values of D for the five species as well as the mixture Mb l - M b 5 when compared to each other ranged from 1.5 to 4.7 x 10 2, with an average value of 3.0 × 10 2. While these values are significantly lower than the maximum value for relatedness determined by Harris and Teller, conclusions concerning sequential identity must await such analysis. Clearly, the amino acid compositions do not show any significant differences at the level of analyses reported here. DISCUSSION
The partial purification of Paramecium myoglobin was reported by Smith et aL (1962), but full purity of the protein(s) was clearly not obtained. In a still later study, Davis & Steers (1976) reported the partial purification of Paramecium myoglobin and its separation into five distinct electrophoretic species. The present study describes the full purification together with several of the physical and chemical properties for the five electrophoretic forms of Paramecium myoglobin. The method described in this study utilizing electrophoretic techniques results in an overall purification of approx 500-fold. Large-scale purification of Paramecium myoglobin is limited by the in vivo levels of the protein and to the maximum volume of culture that can be successfully handled in growing the cells. In the present study, 25 cm 3 of packed cells harvested from 801. of culture fluid yielded an estimated total of 3.6 mg of myoglobin or approx 45/Lg of total myoglobin per liter of culture. A great variability in the myoglobin concentration from culture to culture has been observed, for which no basis has been obvious to date. The total myoglobin yield from harvest to harvest seldom exceeds 4mg/801. culture and drops as low as 1 mg/oulture lot. The conditions employed for growing Paramecium limit the maximum volume of cultures to between 160 and 2001. which can be expected to yield myoglobin levels of approx 2-8 mg/ harvest. Such low levels of material, when taken together with the close isoelectric points of the five molecular forms, prompted our use of electrophoresis as a final step in purification. The visual identification of the pigmented bands in either of the electrophoretic techniques employed here with their near equidistant spacing resulted in the relative ease of eluting highly purified fractions. The molecular weight determinations on the isolated species showed no significant differences. Moreover, mixtures of the various electrophoretic forms gave straight-line plots for log C vs XZ(cm2), indicative of homogeneity of the preparations under conditions similar to those used in electrophoresis. The average molecular weight value calculated for all determinations by sedimentation equilibrium was 15,000 + 500, which agrees closely with the previously determined value of 16,000 obtained by SDS-gel electrophoresis (Davis & Steers, 1976). The amino acid composition does not appear to reflect any significant differences for the 17 residues analyzed. The spectral properties are very similar to those of hemoglobins and myoglobins characterized to date which lead to the conclusion that the five electro-
402
El)WARD STEERSJR and RI('HARD H. DAVISJR
phoretic species isolated from P. tetraurelia are monoheme, single polypeptide chains of approx 15,000 molecular weight. The only definitive differences observed to date among the five forms is in the relative amount in which they are isolated and in their isoelectric points. The values, previously determined by us (Davis & o/ respectively, Steers, 1976), are 32, 11, 12, 33 and 12/,,, for Mb l - M b 5 with isoelectric points of 5.54, 5.18, 4.98, 4.51 and 4.16. The quantitative differences between the five species (Mb 1 Mb 5) has proven to be consistent for several isolations suggesting an in vivo difference. The amino acid composition of the five proteins was analyzed for compositional relatedness by the method of Harris & Teller (1973) with no significant differences evident between the five species. The molecular weight studies and amino acid analyses suggest that the myoglobin species in Paramecium represent only electrophoretic variants. However, this method of analysis is not sensitive enough to detect small differences in primary sequence which may exist and thereby escape detection by the techniques used in this study. In addition, the possible biological significance of variant forms of the myoglobin is not evident. The exact function of these heme-proteins is still unknown. Smith et al. (1962) concluded that "long term" storage function was not likely based on calculations of oxygen consumption and cell numbers. By definition, the heine-proteins of P. tetraurelia are intracellular and therefore not "circulatory" in the usual sense. Earlier suggestions (Smith et al. 1962) that myoglobins (or hemoglobins) with high oxygen affinity may enhance the efficiency of the cytochrome system under conditions of reduced oxygen seem most probable at present. The significance of several electrophoretic species with presumably the same function is even less well understood. Such questions of the possible biological function and significance of the multiple forms of myoglobin in Paramecium await further studies which are now possible with a procedure to isolate the forms in a highly purified state. Acknowledgement--The authors would like to acknowledge the special technical assistance of Mr Clifford E. Lee
in the isolation, cultivation and harvesting of cell cultures and various other aspects of this work. REFERENCES
ANTON1NI E. & BRUNORI M. (1971) Hemoglobin and Myoglobin in their Reactions witl7 Ligands. Frontiers of Biology, Vol. 21 Elsevier, Amsterdam. DAvis R. H. and STEERS E. JR (1976) Myoglobin from the ciliate protozoan Paramecium aurelia. Comp. Biochem. Physiol. 54B, 141 143. HAUT A., Tel)HOPE G. R., CARTWRIGHTG. E. & WINTROBE M. M. (1962) The nonhemoglobin erythrocytic proteins studied by electrophoresis on starch gel. J. clin. hlvest. 41, 57%587. KE1LIN D. & RYLEYJ. F. (1953) Haemoglobin in protozoa. Nature, LoAd. 172, 45l. LOWRY O, H., ROSEBROUGHN. J., FARR A. L. & RANDALL R. J. (1951). Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275. PREER J. R. & PREER L. B. (1959) Gel diffusion studies on the antigens of isolated cellular components of Parameciunl. J. Protozool. 6, 88 100. PROSSI-R C. L. and BROWN F. A., JR (1961) Comparative Animal Physiology. W. B. Saunders, Philadelphia. READ K. R. H. (1968) The myoglobins of the gastropod molluscs, Busycon contrarium Conrad, Lunatia heros, Say, Littorina littorea, L. and Siphonaria gigas, Sowerby. Comp. Biochem. Biophys. 25, 81-94. SATOT. & TAMIYAH. (19371 Cytologia, Fujii Jubilaei Voluman, pp. 1133 1137. SMITH M. H., GEORGE P. & PREliR J. R., JR (1962) Preliminary observations on isolated Paramecium hemoglobin. Arehs Biochem. Biophys. 99, 313-318. TERWlLLIGER R. C. & REAl) K. R. H. (1969) Quaternary structure of the radular muscle myoglobin of the gastropod mollusc Buccinium undatum L. Comp. Biochem. Physiol. 31, 55 64. TERWJLLIGER R. C. & READ K. R. H. (1970) The radular muscle of the amphineuran molluscs, Katharina tunicata, Wood, Cryptochiton stelleri, Middendorf, and Mopalia muscosa, Gould. Int. J. Biochem. I, 281-291. VAN ASSENDELETO. W. (1970) Spectrophotometry of Hemoglobin Derivatives. Thomas, Netherlands. VERNON L. P. & KAMENM. D. (1954) Hematin compounds in photosynthetic bacteria. J. biol. Chem. 211, 643-662. WITTENBERGB. A., OKAZAKIT. & WITTENBERGJ. B. (1965) The hemoglobin of ascaris perienteric fluid. I. Purification and spectra. Biochem. biophys. Acta III, 485-495. YPHANTIS D. A. (1964) Equilibrium ultracentrifugation of dilute solutions. Biochemistry 3, 297-317.
