Studies on iron metabolism in Neurospora crassa

Studies

on Iron

Metabolism

in Neurospora

G. PADMANABAN Department

of Biochemistry,

AND

Indian

Institute

Crassa

P. S. SARMA of

Science, Bangalore,

India

Received December 21, 1964 Neurospora crassa Em 5297a secretes an iron-binding compound (X) when grown under conditions of iron deficiency. Decreasing the concentration of iron in the medium results in an increase of X and a corresponding fall in catalase activity. Under iron-deficient conditions the production of X precedes the fall in catalase activity. The iron complex of the iron-binding compound (XFe) can act as a good iron source to the organism to maintain normal growth and catalase activity, even though the iron is held very firmly in the chemical sense. While ferrichrome is as potent as XFe, as an iron source to N. crassa, ferrichrome A and ferric acethydroxamate are only partially beneficial. XFe, when provided as the sole iron source, also influences nonheme iron enzyme activities like succinic dehydrogenase and aconitase. XFe is permeable to N. crassa mycelia and is incorporated at a much faster rate compared with that from a simple chelate such as ferric citrate.

The isolation of a large number of ironbinding or iron-containing compounds, such as ferrichrome, ferrichrome-A ferrioxamines, Terregens factor, and others including the sideromycins (1, 2) from microorganisms, has led to investigations relating to the transport mechanisms involved in iron metabolism. Furthermore, detailed studies with ferrichrome, isolated from the smut fungus Ustilago sphaerogena(3), have established that this organic iron can act as an iron donor for heme synthesis. It has been demonstrated that ferrichrome and related compounds can act as growth factors for Arthrobacter JG9, Pilobolus kleinii, and for certain other organisms whose requirement can be satisfied by a comparatively high amount of hemin (1). Subsequently, it has been shown in the case of Arthrobacter JG9 that it, has an obligatory requirement for ferrichrome in order to maintain normal growth and catalase activity, even though inorganic iron may be provided in the medium (4). Burnham (5) has observed that ferrichrome-Fe5g can be incorporated into catalase and that this process can be repressed by the addition of hemin in growing cultures of Arthrobacter JG9. He has 147

further demonstrated that cell-free extracts of Rhodopseudomonasspheroidesare able to synthesize hemin when incubated with an oxidizable substrate, protoporphyrin IX, and iron provided as ferrichrome (6). The striking evidence that these compounds exert a role in iron metabolism has been the capacity of several microorganisms to secrete specific iron-binding compounds into the culture fluid when grown under conditions of iron deficiency (1, 7). The iron complexes of several of these compounds have basic similarities with the ferrichromes such as possessinga hydroxamate structure at the iron-binding site and exhibiting a mutual replaceability of one with another as a growth factor for certain microorganisms, despite a difference in over-all structure. It was reported that a new iron-binding compound was isolated from iron-deficient or cobalt toxic cultures of N. crassa and that the new compound was differentiated from ferrichrome, ferrichrome-A, and others on the basis of chromatographic mobilities, solubility properties, and amino acids released on acid hydrolysis of the complex. However, like ferrichrome, the compound has a strong binding affinity for ferric iron

148

PADMANABAN

and binds ferrous iron, if at all, very weakly (8). Evidence has also been obtained as to the presence of a hydroxamate structure (G. Padmanaban and P. S. Sarma, unpublished data). Metabolic studies with ferrichrome have been confined mostly to an organism like Arthrobacter JG9, which has no capacity to synthesize ferrichrome type compounds but responds strikingly to extraneously added ferrichrome. But these compounds primarily should prove metabolically useful in the iron metabolism of parent organisms producing them. The results obtained in this context with N. crassa are presented in this paper. EXPERIMENTAL A wild strain of N. crassa Em 5297a was used. The maintainence of the organism and culture conditions have been described by Sivarama Sastry et al. (9). Briefly, the organism was grown in 10 ml basal medium in 50.ml Pyrex conical flasks in stationary cultures at pH 4.8. The composition of the medium was (gm 100 ml) : glucose, 2; KHzPOI, 0.3; NH4 NOa, 0.2; ammonium tartrate, 0.1; MgS04.7Hz0, 0.05; NaCI, 0.01; CaCIZ , 0.01. Trace elements included are (rg/lOO ml): zinc, 20; manganese, 20; copper, 8; iron, 2; molybdenum, 2. Biotin is added to give a final concentration of 0.5 rg/lOO ml. Preparation of iron-de$cient media. Glucose (A.R.) was rendered metal-free by shaking with Dowex-50 H+ resin and washing the resin with water to free it of the sugar. The major salts (KHzPOI , NHINO, , and ammonium tartrate) were treated with 8-hydroxyquinoline to remove traces of iron. The medium is constituted in metal-free water, omitting iron, and all the other salts used are of analytical grade. When inoculated with a spore suspension from normal slants, this medium permitted about 50yo of the growth obtained in that supplemented with the optimal amount of iron. Isolation of the iron-binding compound from N. crassa. The iron-binding compound (X) was isolated as the iron complex (XFe) from irondeficient culture fluid of N. crassa (8). The steps involve addition of FeC13 , (NH,) zSO~ saturation, extraction of (NHSzS04 supernatant with benzyl alcohol, and re-extraction of the complex into the aqueous phase. The final preparation obtained is chromotagraphically and electrophoretically pure and contains all the iron in bound form. The relative production of the iron-binding compound under different conditions is assessed according to the method of Neilands (1) by adding 1 ml of iron solution (1 mg per milliliter) to 3 ml culture

AND

SARMA

fluid and supernatant

measuring obtained

the optical density after centrifugation

of the at 440

w. FeS9 citrate and XFeSg uptake studies by irondelicient mycelia. The iron-binding compound was isolated as an Fe59 labeled complex (XFe59) and used for uptake studies. Fe59-citrate and XFeS” are added to 40-hour-old iron-deficient cultures of aV. crassa at 10 rg Fe/IO ml medium level in 0.1 ml asceptically. The flasks are then transferred to a reciprocal shaker, and at periodic intervals myCelia and the culture fluid are removed to assess radioactivity and disappearance of XFeS9 from the media, respectively. The radioactivity left over in the medium is also assessed. The mycelia are washed free of adhering radioactivity, dried in an oven at 60°C overnight, and then digested with acid. The acid digests are used to measure radioactivity with a DSS-5 scintillation detector attached to a decade scaler (type 151 A, Nuclear Chicago Corporation, Des Plaines, Illinois). The disappearance of XFes9 from the media is calculated by measuring the optical density of the culture fluid at 440 mp, since XFes9 exhibits maximum absorption at this wavelength. Preparation of enzyme extracts. The mycelia grown for the required length of time are washed with ice-cold water to free them of adhering media, gently pressed dry between folds of filter paper, and then ground in a mortar along with glass powder and phosphate buffer (pH 7.0; 0.05 ill). The extract is then centrifuged at 3000 g for 15 minutes, the precipitate is washed once with the buffer, and the supernatants are pooled. Aliquots of this extract have been used for enzyme assay. Asscly of the enzymes. Catalase was assayed in the myceiial extracts by estimating the amount of HzOa consumed with a permanganate titration method as described by Ramachandran and Sarma (10). Succinic dehydrogenase was assayed in the same extracts by t.he decolorization of dichlorophenol indophenol in the presence of cyanide to inhibit cytochrome oxidase; the method used is essentially that of Green et al (11). Aconitase activity was measured in phosphate buffer extracts by following the hydrolysis of cisaconitic anhydride (12). Citric acid was estimated as described by Stern (13). In some experiments where the in vitro effects of iron and cysteine on the enzyme have been studied, cold water extracts were used instead of the phosphate buffer extracts. Protein contents were measured by the method of Lowry et al. (14). RESULTS

The iron-binding compound was detected in the culture fluid of N. crassa only under conditions of iron deficiency, whether direct

IRON

METABOLISM

.IN.

or conditioned due to cobalt toxicity (8). Table I indicates that the production of the iron-binding compound progressively increases with a decrease in the iron status of the medium from the optimal level, whereas the catalase activity shows a corresponding decrease. When the production of the ironbinding compound and catalase activity are assayed as a function of the growth period when the organism is grown under iron-deficient conditions, it was found that the iron-binding compound is secreted into the medium even at 24 hours growth, xv-hen over-all growth and catalase activity are not affected (Table II). The data presented in Table III indicate that XFe and ferrichrome, when provided as iron supplements to the basic iron-deficient medium, enable the organism to maintain normal growth and catalase activity. Ferrichrome A and ferric acethydroxamate are only partially beneficial. It TABLE

svora Iron

ACTIVITY,

I

TABLE

AND

EFFECT FERRIC

OF Neuro-

GROWTH

Iron-binding compound produced o.d. at 440 mp

Cat&se activity, ml O&fmIyy;2 protein/S min

0.24 0.26 0.22 0.20 0.11 0.03

8.0 10.1 12.7 19.3 31.2 54.0

0.05 0.10 0.20 0.50 1.00

Compound

(ln$r;t:.t.,

OF IRON

ACTIVITY,

DEFICIENCY AND GROWTH

ON THE

24 48 72

Iron-binding compound produced; o.d. at 440 mp

0.03

OF THE

OF GROWTH

IRON-BINDING

PERIOD

media

6.1 42.7 55.0

Catalase activity, ml 0.01 M KMnOa consumed/mg protein/j min

24.5 40.4 42.5 32.8 35.0 44.4

13.4 53.9 55.6 33.8 37.0 56.1

COMPOUND,

IN Neurospora Iron-deficient

Cat&se activity, ml O.OlM KMn04 consumed/mg protein/S min

6.0 30.5 42.5

OF

Growth (m~,$ry

of 1 pg experi-

II

PRODUCTION

AS A FUNCTION

Normal Period (hours)

addeda

ACTIVITY

(1 The compounds were added at a level iron/l0 ml iron-deficient medium. The mental details are given in text.

to a basic details

TABLE EFFECT

FERRICHROME A, AND IXORGANIC

CATALASE

AND

Nil XFe Ferrichrome Ferrichrome A Ferric acethydroxamate FeC13.6Hz0

22.8 24.6 28.6 32.G 39.0 45.0

a Graded levels of iron were added iron-deficient media. The experimental are given in text.

III

OF XFe, FERRICHROME, ACETHYDROXAMATE,

IRON ON GROWTH Neurospora crassa

crassa

added (PgY

149

cra.s~a

is clear that the iron of XFe is available to the organism, even though it is held very strongly in the chemical sense (8). When XFe is provided as the sole iron source, by adding-it to a 40-hour-old irondeficient culture (by which time iron deficiency has set in significantly), it can influence not only catalase activity and the final growth reached but also the activities of non-heme iron enzymes like aconitase and succinic dehydrogenase. These results are presented in Table IV. Since evidence is available that the aconitase enzyme can be activated by preincubation with iron and cysteine (12), the N. crass-a crude extract, from normal and iron-deficient mycelia, was also preincubated with Fe* and cysteine as well as XI’e and cysteine. In both the cases no activation was detectable. In this context the recent work of Palmer (15) on plant aconitase is of interest since he failed to obtain any in

EFFECT OF IRON STATUS OF THE MEDICM ON THE PRODUCTION OF THE IRON-BINDING COMPOUND, CATALASE

N.

CATALASE crassa

media

Iron-binding compound produced; o.d. at440lnP

Cat&se activity, ml O.OlM KMnO, consumed/mg protein/S min

Growth (mg dry wt.)

0.04 0.10 0.24

7.0 8.5 10.2

5.9 19.5 24.0

150

PADMANABAN

AND TABLE

EFFECT

OF INORGANIC GROWTH

ON

AND XFe SUCCINIC

IRON CATALASE,

0~

40-hour normal 40-hour iron deficient 72-hour normal 72-hour iron deficient XFe added at 40 hours Iron added at 40 hours

iVeurospora

100 (38.5) 25 138 30 110 119

TABLE

XFP

0 15 30 45 60 120

1.76 1.85 1.83 1.89 1.91

uptake

2.29 0.44 0.35 0.37 0.32 0.29

HOURS

Aconitase activity, Irg citric acid/mg protein/E min

100 (0.10) 70 120 40 92 87

BY do-HOUR-OLD

UPTAKE

AT 40 GROWTH

crassa”

Succinic dehydrogenase activity, fall in o.d. at 600 m&ng protein/3 min

a The additions were made at a level of 1 rg Fe/l0 expressed as percentages of the values recorded for The actual values are given in parentheses.

XFe5Q AND Fe69-CITRATE

IV

ADDED TO IRON-DEFICIENT CULTURES DEHYDROGENASE, ACONITASE, AND

Cat&we activity, ml O.OlM NM”04 consumed/mg protein/S min

Treatment

SARMA

100 (97.0) 84 70 40 62 60

ml medium. The results 40.hour normal mycelia,

25.2 18.0 46.0 26.0 42.0 44.0

of ensymic activities are which are taken as 100.

V IRON-DEFICIENT FP-citrate

MYCELIA uptake Ri;dEs;Fty

OF Neurospora

crawa

Fe’g-citrate uptake in presence of iron-free compound

o.d. medium at 440 nw

Radioactivity in mycelium, counts/min/ mycelium (X 10’)

counts/n& 10 ml medium (X 10’)

Radioactivity in mycelium, counts/min/ mycelium (X lo”)

Radioactivity in medium, count&in/ 10 ml medium (X 109

0.130 0.010 0.008 0.008 0.008 0.008

0.40 0.72 0.99 1.25 1.86

2.29 1.87 1.53 1.28 0.95 0.35

1.75 1.81 1.81 1.85 1.86

2.29 0.44 0.40 0.42 0.30 0.36

u The additions were made at 10 rg Fe/l0 ml medium in 0.1 ml volume to 40-hour-old iron-deficient cultures. b From the radioactivity actually detected in the mycelium and that left over in medium, the losses due to washing of mycelium and other manipulations have been assessed. The loss is about 5-7yo of the actual radioactivity expected on the mycelium (calculated from the fall of radioactivity in the medium) and is in the same range in all cases.

activation with the purified enzyme, even though iron deficiency in the plant is known to result in decreased levels of the enzyme. In the present study it was also found that XFe has no in vitro activating effect on catalase or succinic dehydragenase activities. Since the metabolic potency of XFe as an iron donor has been established, the interest had turned to whether XFe is permeable to the cell or the iron is split off extracellularly and then incorporated. XFe5g was provided at 10 pg Fe/l0 ml medium to the 40-hour-old, iron-deficient cultures, and the flasks were shaken in a reciprocal shaker. vitro

At different intervals of time, the incorporation of radioactivity into the mycelium and the disappearance of XFe from the medium were assessed by measuring the fall in optical density at 440 rnp (XFe has maximum absorption at 440 mp). Similar experiments were also conducted with iron given as a simple chelate such as Fesg-citrate. The results presented in Table V indicate that, whereas XFe59 incorporation reaches a maximum within 15 minutes, Fe5g-citrate takes nearly 2 hours to reach a similar level of incorporation. When iron is removed from XFe by alkali treatment (8) and the ironfree fragment is added to Fe5g-citrate, a

IRON

METABOLISM

IN

N.

cras~a

151

similar enhanced rate of FeSg incorporation exchanged when reincubated in a fresh is again observed as compared with that medium containing cold XFe or inorganic when Fe5g-citrate alone is used. The mycelial iron. weights remain more or less a constant DISCUSSION throughout the incubation period. Fe69C13 The production of the iron-binding comincorporation follows the same rate as that pound in N. crassa is strikingly dependent on of Fesg-citrate. the iron concentration in the medium, as is To further substantiate that XFe was the case with the production of the other taken in as an intact molecule, the mycelia iron-binding compounds reported by were washed well just after 5 minutes incubation with XFe5g, by which time the Fe5g Neilands (1). A progressive increase in the production of the iron-binding compound taken in would not have been dissipated and a corresponding decrease in catalase to other systems appreciably, and were levels with the fall in the iron status of the extracted with phosphate buffer; the buffer growth medium establishes the dependency extract was then processed as for XFe isoof both the systems on iron nutrition and a lation (8). The final preparation chromatreciprocal relationship between the two. ographed on paper in two different solvent Ferrichrome is the natural iron-containing systems gave a radioactive spot correspondof Ustilago sphaerogena, and ing to XFe and accounting for 60 % of the metabolite ferrichrome A is the iron-binding moiety total iron uptake. Studies with identical mycelia provided with Fe5g-citrate and secreted under conditions of iron deficiency by the same organism (3, 16). Burnham processed under identical conditions indiand Neilands (4), on screening several cate that the spot corresponding to XFe microorganisms for ferrichrome activity, can account only for 10% of the total radio(16117) gives a strongly activity incorporated. This 10% may be found that N. C~CLSSCL due to the formation of the complex in the positive reaction and answers as well the It is highly mycelia as well as the small amount of the test for bound hydroxylamine. iron-binding compound already present in probable that N. crassa Em 5297a also synthesizes its own organic iron component the 40-hour iron-deficient culture fluid. (X’Fe) with a secondary hydroxamate These results are presented in Table VI. structure corresponding to the ferrichrome of Within the short time (15 minutes) of U. sphaerogena, when grown with normal maximal XFe59 incorporation and disaplevels of iron in the medium. The ironpearance from the medium, no possible breakdown products could be detected and binding moiety (X) secreted under condithe radioactivity incorporated could not be tions of iron deficiency of N. crassa may be the same as X’, or the two may be closely related. TABLE VI If X’Fe were to act as the natural iron DETECTION OF XFeb9 IN THE MYCELIUM~ donor for heme synthesis by N. crassa, the Radioactivity in XFe spot primary site to be affected in iron deficiency Radioactivity in mycelium Butanol:acetic Methanol: CoFdyeyd would be the formation of the iron-chelate (countsjminj acid: water; water; mycelium) Rf = 0.56 Rf = 0.88 before there is a fall in heme synthesis. (counts/min) (counts/min) Naturally, the secretion of the iron-binding moiety would precede the fall in catalase XFeS9 6080 3600 3480 Fesg-citactivity when the organism is grown under 1560 150 175 rate iron deficient conditions (Table II). The survival value of the phenomenon of a 40-hour-old iron-deficient mycelia after inthe iron-binding compound production to cubation with XFe69 and FeQitrate, added at the parent organism is established by the 10 rg Fe/l0 ml medium, were processed after metabolic potency of XFe as an iron donor thorough washing to remove adhering radioacto maintain normal growth and catalase tivity, and the final preparation was chromatoactivity, even though the iron is held very graphed on paper in two solvent systems. The strongly in the chemical sense. This is radioactivity added has been adjusted to give 2.29 X 10” counts/min/lO ml medium. further emphasized by a certain specificity

152

PADMANABAN

shown by the organism as regards the nature of the organic iron it can utilize. Whereas ferrichrome and XFe serve as good sources of iron, ferrichrome A and ferric acethydroxamate do not serve equally well, when all these are provided at the same iron level. The nutritional inactivity of ferrichrome A has already been noted with Arthrobacter JG9 (4), and it should be of exclusive importance only to the parent organism U. sphaerogena. Earlier work implicating ferrichrome in iron sequestration and transport has been mainly confined to the role of these compounds in supplying iron for heme synthesis. Burnham (5) has envisaged a possibility that in Arthrobacter JG9 there can be two routes for iron metabolism, one to the heme through the ferrichrome and the other representing the nonheme iron. But studies with XFe in the present investigation have revealed that when it is provided as the sole iron source, it not only can support normal growth and catalase activity but can also influence the levels of nonheme iron enzymes like suceinic dehydrogenase and aconitase. While little is known regarding the active center of these enzymes, and the role of iron in influencing the activities of these enzymes may be indirect, it is clear that XFe can supply iron for such a role. At least in the parent organism N. crassa, XFe can assume a generalized role, controlling iron supply to heme as well as nonheme iron enzyme systems. Even in Arthrobacter JG9, where it has been established that ferrichrome influences heme synthesis, the effect of ferrichrome deprivation on nonheme iron enzyme systems may give rise to interesting results. The last evidence cited to indicate the metabolic potency of XFe is the avidity with which the iron-deficient N. crassa mycelium takes up the compound as compared with the rate of entry of iron at rate or inorganic iron. Whether this is due to a special mechanism evolved by the organism for a preferential permeability of the complex is yet to be determined. A parallel situation also exists in tomato plants where ferrioxamin B translocation to the upper parts of the plant takes place more rapidly than ionic iron (17). Finally, we must await

AND

SARMA

further investigation on the isolation of the natural metabolite X’Fe from the mycelia of N. crassa when grown under normal conditions with optimal levels of inorganic iron, and its participation in the iron metabolism of this organism. Such a study is warranted by the metabolic potency of XFe when isolated under conditions of iron deficiency from the culture fluid. ACKNOWLEDGMENT Ferrichrome and ferrichrome A were obtained through the kind courtesy of Dr. J. B. Neilands. Thanks are due to Mr. T. S. Satyanarayana for technical assistance. The financial assistance from the Council of Scientific and Industrial Research, New Delhi, and the Rockfeller Foundation, New York, is gratefully acknowledged. REFERENCES 1. NEILANDS, J. B., Bacterial. Rev. 21, 101 (1957). 2. BICKEL, H., GBUMANN, E., KELLER-SCHIERLEIN, W., PRELOG, E., VISHER, E., WETTESTEIN, A., AND ZHHNER, H., Experientia 16, 129 (1960). 3. NEILANDS, J. B., J. Am. Chem. Sot. 74, 4846 (1952). 4. BURNHAM, B. F., AND NEILANDS, J. B., J. Biol. Chem. 236,554 (1961). 5. BURNHAM, B. F., Arch. Biochem. Biophys. 97, 329 (1962). 6. BURNHAM, B. F., J. Gen. Microbial. 32, 117 (1963). 7. ZBHNER, H., KELLER-SCHIERLEIN, W., Hii~TER, R., HESS-LEISINGER, K., AND DQER, A., Arch. Mikrobiol. 46, 119 (1963). 8. PADMANABAN, G., AND SARMA, P. S., Arch. Biochem. Biophys. 108, 362 (1964). 9. SIVARAMA SASTRY, K., ADIGA, P. R., VENKATASUBRAMANIAM, V., AND SARMA, P. S., Biochem. J. 86, 486 (1962). 10. RAMACHANDRAN, L. K., AND SARMA, P. S., J. Sci. Ind. Res. 13l3, 115 (1954). 11. GREEN, D. E., MII, S., AND KOHOUT, P. M., J. Biol. Chem. 217, 551 (1955). 12. MORRISON, J. F., Biochem. J. 66,99 (1954). 13. STERN, J. R., in “Methods in Enzymology” (S. P. Colowick and N. 0. Kaplan, eds.), Vol. 3, p. 425. Academic Press, New York (1957). 14. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. I,., AND RANDALL, R. J., J. Biol. Chem. 193, 265 (1951). 15. PALMER, M. J., Biochem. J. 92,404 (1964). 16. GARIBALDI, J. A., AND NEILANDS, J. B., J. Am. Chem. Sot. 77,2429 (1955). 17. STUTZ, E., Experientia 20,430 (1964).