Biochimica et Biophysica Acta. 990 (1989) 45-52Elsevier

BBA 23024

Sedimentation properties of chitosomal chitin synthetase from thewild-type strain and the 'slime' variant of Neurospora crassa

Jose P. Martinez, Gloria Gimenez and Salomon Bartnicki-GarciaDepartamento de Microbiologia, Facultad de Farmacia, Uniuersidad de Valencia, Valencia (Spain)

(Received 29 July 1988)

Key words: Chitin synthetase; Sedimentation property; Chitosome; Isopycnic sedimentation; (N. crassa'y

45

Marked differences in the pattern of sedimentation of cellular structures were observed after isopycnic centrifugation ofcrude cell-free preparations from the Neurospora crassa wall-less 'slime' variant and mycelial wild-type strain. Kineticstudies of particle sedimentation showed that the various types of subcellular components, as revealed by turbidity, UVabsorption, polypeptide patterns, and chitin synthetase activity determinations, sediment independently of one another.An important feature was the finding that chitin synthetase from 'slime' peaked at a median specific gravity of1.1201 ± OJ)036, whereas that from wild-type strain sedimented at a higher buoyant density (specific gravity 1.1349 ±0.0024). Different cultivation conditions or cell breakage procedures (osmotic lysis or ballistic disruption) did not seemto affect this sedimentation behavior. Electron microscopy revealed the presence of chitosomes (microvesiclescontaining chitin synthetase) in the chitin synthetase activity peaks obtained after isopycnic centrifugation of cell-freeextracts from 'slime' and wild-type strains. The discrepancy in buoyant density of chitin synthetases from both N.crassa strains might point to inherent differences in chemical composition of the chitosomal microvesicles. In any case,the lower buoyant density of 'slime' chitosomes appears to be one of several major alterations in sedimentation behaviorof subcellular structures. These alterations might be related to the inability of 'slime' to make a cell wall.

Introduction

Chitin synthetase (UDP-N-acetyl-o-glucosamine:chitin 4-,B-N-acetylglucosarninyltransferase, EC 2.4.1.16)has been isolated from all major groups of chitinousfungi in the form of microvesicular organelles calledchitosomes that can synthesize chitin microfibrils 'invitro' [1-4]. It is believed that chitosomes are cyto­plasmic conveyors of zymogenic chitin synthetase, whichtransport the enzyme to its final destination at the cellsurface where chitin microfibrils are synthesized andassembled [5]. Chitosomes were first discovered in theyeast form of Mucor rouxii [2,4], and were describedlater in other chitinous fungi including Neurosporacrassa (5].

Wild-type N. crassa grows as a mycelium with typi­cal hyphal walls, but a variant of this species called

Abbreviations: SDS-PAGE, sodium dodecyl sulphate-polyacrylamidegel electrophoresis; UDP-GIcNAc, undine diphosphate N-acetylglu­cosarnine; G1cNAc. N-acetylglucosamine; d. buoyant density ing/cm3•

Correspondence: J.P. Martinez, Departamento de Microbiologia,Facultad de Farmacia, Universidad de Valencia, Avda. Blasco Ibanez13, 46010 Valencia. Spain.

'slime' lacks cell walls and grows as spheroidal cells.The 'slime' phenotype is believed to be the result of atleast three different morphological mutations desig­nated fuzzy (fz), osmotic (os), and spontaneousgermination (sg) [6,7]. Although it has been postulatedthat expression of the sg mutation might be responsiblefor the absence of cell wall material in 'slime', since itwould cause inhibition of chitin synthesis [7], this hy­pothesis has not yet been confirmed.

Previous comparative studies on chitin biosynthesisbetween the wild-type strain and 'slime' variant of N.crassa revealed no reason for the inability of 'slime' tomake chitin. Thus, chitin synthetases isolated from bothstrains were found to have similar catalytic and kineticproperties [8], zymogenic character, and to be mainlylocated in chitosomes [5,9]. Substrate availability is nota likely explanation because there does not seem to be alack of substrate (UDP-GlcNAc) [7,10], or an obviousphysical barrier to its utilization, since UDP-GlcNAc islocated in the cytosol of both 'slime' and wild-type [111.

Low buoyant density facilitates separation of chi to­somes of M. rouxii from other membranous structuresof a cell homogenate [12]. A preceding study on thechitosomes of the' slime' mutant of N. crassa showedbuoyant density values that were considerably lower

0304-4165/89/$03.50 (0 1989 Elsevier Science Publishers B.V. (Biomedical Division)

46

than those for M. rouxii chitosomes [9]. Thus, we de­cided to investigate whether the exceedingly low buoyantdensity was a general property of chitosomes of N.crassa or a peculiarity of the unique wall-less 'slime'variant.

Materials and Methods

Strains and culture conditionsN. crassa wild-type strain FGSC 988 (St. Lawrence

74-0R8-1a) and the wall-less 'slime' variant FGSC 1118were used throughout this work. Both organisms wereobtained from the Fungal Genetics Stock Center(Arcata, CA).

The following culture media were employed. MediumA: Vogel's N medium [13] supplemented with 2% (w/v)sucrose. Medium B: the complete medium described byBeadle and Tatum [14] except that the alkali-hydrolyzedyeast nucleic acid was omitted, and Difco malt extractwas used instead of spray-dried malt syrup. Medium C:Scarborough's 'diet' medium [15]. Medium D: a modifi­cation of Scarborough's 'diet' medium with the follow­ing composition: Vogel's N medium supplemented with2% (w/v) sucrose, 7.5% (w/v) sorbitol, 0.75% (w/v)yeast extract, and 0.75% (w/v) nutrient broth. Thewild-type strain was maintained on 2% (w/v) agarslants of medium A or B. The •slime' variant wasmaintained on 1.5% (w/v) agar slants of medium C orD and subcultured every 15 days. All slant cultureswere incubated at 28 0 C.

Mycelium of wild-type N. crassa was propagated inliquid cultures. Erlenmeyer flasks (2 liter) containing500 ml of either liquid medium B or D described abovewere inoculated with (1-1.2)· 106 conidiayrnl with aconidium suspension in distilled water obtained fromcultures grown for 10-12 days on solid medium B or Arespectively. The liquid cultures were incubated at 28 0 Cfor 13-15 h in a New Brunswick orbital shaker (230-250rpm). 'Slime' cells were also grown in two differentliquid media as follows: 250-ml flasks containing 50 mlof medium C or D were inoculated with a loopful of'slime' cells from slant cultures grown for 8-10 days(each liquid medium was inoculated with cells previ­ously grown on agar slants of the same culture medium).The flasks were kept still at 28 0 C for 24 h and thenincubated at the same temperature on a New Brunswickreciprocal shaker (90-100 strokes) for 24 additionalhours. Fifty milliliters of this initial culture were addedto 450 ml of each fresh medium and incubated withshaking for 24 h as above. At harvest time, the ab­sorbance at 650 nm (A 6S0 ) of the cultures was in therange of 0.7-1.0.

Isopycnic sedimentation of chitosomal chitin synthetasefrom 'slime' variant

All steps were carried out at 0_4 0 C. Cells from two

500-ml cultures (average A 6 S0 = 0.85) were harvested bycentrifugation (600-700 X g, 10 min, in 500 ml Nalgenebottles) in a Damon/IEC CRUSOOO centrifuge equippedwith a swinging bucket head, washed once with sorbi­tol/phosphate buffer (33 mM NaH2P04-NaOH, pH8.2 plus 0.7 M sorbitol) and then pelleted by centrifuga­tion at the same speed. The resulting cell pellet (about3.5 rnl) was divided into two portions, and the cellsbroken by two different procedures:

(A) Ballistic disruption: 1.5 ml of cell pellet wasplaced in a 75 ml Duran sample flask together with 10ml of 0.45-0.5 mm diameter glass beads, and 15 ml ofphosphate buffer (33 mM NaH2P04-NaOH, pH 8.2).The suspension was shaken in a Braun MSK homo­genizer for 20 s. The flask was cooled with liquid CO2

during cell disruption. This procedure broke all cells.The homogenate was removed by decantation and thebeads were washed once with 1 ml of phosphate buffer.Homogenate and washings were combined, and thefinal volume was adjusted to about 18 ml with phos­phate buffer.

(B) Osmotic lysis: 1.5 ml of cell pellet was dilutedwith 15 ml of phosphate buffer, and the resulting sus­pension was homogenized by aspiration through anordinary Pasteur pipette. Cell breakage was about 90%.

The cell homogenates were centrifuged at 1000 X gfor 10 min to remove unbroken cells and large celldebris, and the supernatant obtained was used as thecrude cell-free extract. Variable amounts of the cell freeextract (from 0.3 to 1 ml, depending on the chitinsynthetase activity present in the cell free extract) werelayered on top of linear sucrose density gradients(specific activity of chitin synthetase in the samplesloaded onto the gradients was in the range of 0.3 to 0.6unitsj/mg of protein) made in phosphate buffer (12 rnl;10-65% (w/v) sucrose). Magnesium was omitted fromthe buffer in the gradient to avoid precipitation orparticle aggregation, but a final concentration of 10 mMMgCI 2 was always present in the chitin synthetase assaymixtures since magnesium considerably increases theactivity of N. crassa chitin synthetase in the in vitroassays (unpublished).

The gradients were centrifuged at 83000 X g (RaY) ina Beckman SW-41Ti rotor for 19 h. Centrifuged gradi­ents were illuminated overhead and photographed withan MP-4 Polaroid camera. Fractions of 0.5 ml werecollected with an ISCO fractionator. After fractiona­tion, any residue at the bottom of the centrifuge tubeswas resuspended in 0.5 ml of phosphate buffer, andchitin synthetase activity was also measured in thissuspension. The specific gravity of each individual frac­tion was calculated from the sucrose concentration mea­sured with an Abbe refractometer (Carl Zeiss, Ob­erkochen, F.R.G.). Absorbance was continuously moni­tored during fractionation at either 280 or 260 nm withan ISCO UV-detector.

Isopycnic sedimentation o[ chitosomal chitin synthetase[rom wild-type strain

Mycelia from a 500 ml culture were collected byfiltration through Whatman No.1 paper, washed oncewith 50 ml of ice-cold phosphate buffer, and resus­pended in 20 mI of the same buffer. This cell suspensionwas mixed with 20 ml of dry glass beads (0.45-0.5 mmdiameter) and the cells were broken for 45 s with aBraun MSK homogenizer. Breakage of mycelial cellswas complete under these conditions. The homogenatewas centrifuged at 1000 X g for 10 min to eliminate thecell wall fraction, and the resulting supernatant wasused directly as the crude cell free extract, or subse­quently centrifuged at 54000 X g (Ray) for 45 min in aBeckman Type 30 rotor. The 54000 X g supernatantcontained the 'miniorganelle' population of the cell(ribosomes, miscellaneous enzyme particles, and micro­vesicles, including chitosomes) together with solublecomponents of the cytoplasm [12]- Linear sucrose den­sity gradients (12 rnl; 10-65% (w/v» were loaded withvariable volumes (0.3-1 ml) of the 1000 X g or the54000 X g supernatants (specific chitin synthetase activ­ity in the samples loaded onto the gradients was in therange of 0.3-0.5 unitsyrng of protein). Gradients werecentrifuged and manipulated as described above.

Chitin synthetase activityChitin synthetase activity was assayed by the filtra­

tion method described earlier [16]. The standard reac­tion mixture contained 0.65 roM UDP-[14C]GlcNAc(0.22 Cijmol), 26 mM GlcNAc, 0.26 mM ATP, 10 mMMgCI 2 , phosphate buffer, and enzyme in a final volumeof 0.140 ml. In all assays, zymogenic chitin synthetasewas activated with trypsin (final concentration in theassay mixture, 10 p.g/ml). Assay mixtures were in­cubated for 1 h. Activity was expressed as units, 1 unitbeing the amount of enzyme that catalyzes the incorpo­ration of 1 nmol GlcNAc into chitin per min. Specificactivity was related to 1 mg protein.

Analytical polyacrylamide gel electrophoresisAnalytical sodium dodecyl sulfate-polyacrylamide gel

electrophoresis (SDS-PAGE) was performed on 10%(w/v) acrylamide gels as described by Laemmli [17].Samples were treated with a solution of 2 g SDS, 10 gglycerol, and 4 ml 2-mercaptoethanol in 17.5 ml of 0.5M Tris-HCI, pH 6.8. The sample/ solubilizing solutionvolume ratio was 2 : 1. A constant current of 25 rnA perslab was applied until the bromphenol blue tracking dyefront was about 1 ern from the bottom of the resolvinggel (15 rnA constant current was applied through thestacking gel). Proteins were revealed by the silver stain­ing procedure described by Wray et al. [18]. The follow­ing molecular weight standards were run in parallel:lysozyme (14400), soybean trypsin inhibitor (21 500),carbonic anhydrase (31000), ovalbumin (45000), bovineserum albumin (66200), and phosphorylase b (92500).

47

Electron microscopyObservations were made on negatively stained speci­

mens prepared as follows. Droplets of sample (5-10 p.l)were placed on carbon-coated copper grids. After 5min, a volume similar to that of the sample of 2%glutaraldehyde in 0.1 M sodium cacodylate buffer (pH7.1) was added, and the mixture was left to stand for 5more min. After fixation, the grids were rinsed withglass-distilled water to remove glutaraldehyde andsucrose. Specimens were stained with 2% aqueous uranylacetate for 2 min, withdrawn with filter paper, air-dried,and examined and photographed with a JEOL 100Selectron microscope operating at 60 kV.

MiscellaneousProtein was measured by the Lowry method. Uridine

5'-diphosphate N-acetyl-D-[1-14 Cjglucosamine was ob­tained from ICN (Irvine, CA, U.S.A.). All SDS-PAGEreagents and the standard protein electrophoresiscalibration kit were from Bio-Rad (Richmond, CA,U.S.A.). Ultrapure density grade sucrose was purchasedfrom Schwarz-Mann (Orangeburg, NY, U.S.A.). Allother reagents were of analytical grade.

Results

Sedimentation of chitosomal chitin synthetase from 'slime'and wild-type strains

Centrifugation of CFE (1000 Xg supernatants) fromboth strains on 10-65% (w/v) sucrose gradients at83000 X g for 19 h produced one major, sharp, andnearly symmetrical peak of chitin synthetase activity(Fig. lA, B). Examination under the electron micro­scope of samples taken from fractions exhibiting thehighest chitin synthetase activity within each peak re­vealed the presence of microvesicles (Fig. lA, lB, insets)with the typical morphology and size of chitosomesshown previously in N. crassa and other fungi [2,5]. Afew larger vesicles were also present as well as a numberof smaller non-vesicular structures of irregular shape,most of which were probably of ribosomal origin. Therewere no obvious morphological differences between thechitosomes of 'slime' and wild-type N. crassa.

An obvious difference between the chitosomes of'slime' and wild-type was buoyant density. In sevenseparate experiments (Table I), the chitosome peak fromwild-type strain was of significantly higher density thanthe corresponding peak from 'slime' (Fig. 1). To obtaina fair comparison of buoyant densities, instead of usingthe value of specific gravity for the fraction with thehighest chitin synthetase activity, we calculated themedian specific gravity for the entire activity peak(Table I). Accordingly, the median specific gravity val­ues for the chitosomal peak of 'slime' ranged fromd= 1.1137 to 1.1261, the mean of this group of medianvalues being d = 1.1201 g/cm3

• In contrast, the mean of

ooCDC'I

-cII

o

I a

20

1 2

24

20

1 6

00

1 2 co(\j

<{

8

4

0

1.2111.24

c

1.08 1.1 2 1.1 6 1.20

SPECIFIC GRAVITY

A

jJ.o..,d q

\ ,

O l.---L_ l-...--L_ .L.---L_.L.---'-_ .L.---L-JL....--I----I:::.---I----I-J

1

0

9

8B

......c"e 1"::::

0E 6c~

Well 5<l:I-W:r It Rl- I \2 9 \>

3 I qIf)

,f!> \zo '!:: 2 \

:r q,o~-o-•

48

9

8--c"e 7......(5E 6.5well 5<l:I-W:r IiI-Z>ell 3zi= 2i:o

•

Fig. 1. Distribution of chitin synthetase activity after isopycnic sedimentation. One milliliter samples of cell free extracts (1000 X g supernatants)from wild-type strain (A; total chitin synthetase activity loaded, 21 units; specific activity, 0.46 unitsy'mg of protein), and' slime' (B; total activityloaded, 8.56 units; specific activity, 0.32 unitsy'mg of protein), were applied on top of 100-650 g/l linear sucrose gradients and centrifuged at83000x 8 (ROY) for 19 h in a Beckman SW-41Ti rotor. Fractions (0.5 ml) were collected and 30 fIol assayed for chitin synthetase. Bold arrowsindicate specific gravity median values for the entire ch.itin synthetase activity peaks. A 260 values correspond to undiluted samples. The lettersindicate fractions analyzed by SDS-PAGE (see Fig. 2). Electron micrographs in part A and B respectively show chitosomes in fractions c (wild-typestrain), and f (' slime' variant). In each case, white arrowheads point to microvesicles exhibiting chitosomal morphology; fa: fatty acid synthetase

particles. Negatively stained specimens (bar: 100 nm).

the median values for wild-type was d = 1.1349 with arange from d = 1.1307-1.1390 (Table I).

The above-mentioned sedimentation pattern ofchitosomal chitin synthetase from both strains seemsnot to be affected by cultivation conditions, or by thecell homogenization procedure. Thus, when two drasti­cally different breakage methods were compared ­osmotic lysis vs. ballistic disruption - there was noalteration in the sedimentation pattern of chitosomesfrom the 'slime' variant (Table I). Also, growth in

different culture media did not affect the sedimentationproperties of chitin synthetase from both 'slime' andwild-type (Table I).

SDS-PAGE analysis of fractions from the chitin syn­thetase activity peaks of 'slime' and wild-type revealed12 major polypeptides in the wild-type fractions (Fig. 2,lanes a-d), whereas four major bands plus 4-5 faintones were detected in the' slime' variant samples (Fig.2, lanes e-h). Seven polypeptides appeared in bothpatterns but only four of them were common major

49

TABLE I

Equilibrium sedimentation of chitosomal chitin synthetase from 'slime' and wild-type strains of N. crassa

Variable volumes (from 0.3 to 1 rnl) of ceII free extracts (1000 X g or 54000 X g supernatants) obtained from both N. crassa strains, as described inMaterials and Methods. were loaded onto linear sucrose gradients (10-65%, w/v). The gradients were centrifuged at 83000x g (R •• ) for 19 h andfractionated into 0.5 ml aliquots. Chitin synthetase activity and sucrose concentration were determined in each collected fraction.

Culture " CeII b Cell free extract 'Slime' Wild-typemedium disruption peak d median • peak d median •

BT B SI 1.1398(1) 1.1390SD B Sl 1.1209(1) 1.1261MSD B SI 1.1182(0.3) 1.1203 1.1297(0.35) 1.1349MSD B SI 1.1161(0.5) 1.1206 1.1315(0.5) 1.1341MSD B Sl 1.1082(0.65) 1.1137 1.1353(0.5) 1.1355MSD D si 1.1145(0.7) 1.1202MSD B Sl 1.1350(0.35) 1.1350MSD B S54 1.1351(1) 1.1307MSD 0 Sl 1.1113(1) 1.1195

Mean: 1.1149 1.1201 1.1344 1.1349S.D.: 0.0042 0.0036 0.0032 0.0024

• Strains cultivated in: BT, Beadle and Tatum complete medium; SD: Scarborough's 'diet' medium; MSD: modified Scarborough's 'diet' medium(see Material and Methods).

b B: balIistic; 0: osmolysis.c Sl: 1000 x g supernatant; S54: 54000 X g supernatant.d Peak value is the specific gravity in the fraction with the maximum chitin synthetase activity; number in parentheses indicates the volume of the

corresponding ceII free extract preparation subjected to centrifugation.C Median value calculated for the entire peak of chitin synthetase activity (see text).

components in the peak fractions (Fig. 2, lanes c and C,open arrowheads) of 'slime' and wild-type (polypeptideswith molecular masses of 49, 23, 17, and 13 kDa).

pycnic centrifugation of the cell free extracts (1000 X gsupernatants) from wild-type and 'slime' revealedmarked differences in turbidity bands. Most of the

Sedimentation pattern of cell-free extracts of 'slime' andwild-type

Visual inspection of the sucrose gradien ts after iso-

Fig. 3. Distribution of particulate subcellular structures after centrifu­gation of cell free extracts (1000 X g supernatants) from wild-type (1)and 'slime' (2), as described in Fig, L The different turbidity bandsare identified by letters (see text). White arrows point to the locationof the chitin synthetase activity peaks relative to the turbidity bands

observed within the gradients.

h9

1- . ...,- 92 0

~

)(

-66 ~:E

- 1-- - -45--

- 3 1

fe

B · .. · . .

dcb

-- 1- !> ~. __..., ..'-

a

..- !> - - -~ -21- --- ---e- r--

e- - 14

A - _.. . .

Fig. 2. Polypeptide composiuon of fractions comprising the chitinsynthetase activity peaks of wild-type (A) and' slime' (D). Letterscorrespond to fractions in Fi8. 1. 100 1'1 samples of each fraction weremixed with 50 ~I of solubilizing solution (see Materials and Methods),and then boiled in a water bath for 5 min. 100 1'1 from each of thedifferent sample/solubilizing solution mixtures were loaded on 10%slab gels and subjected to electrophoresis. Open arrowheads point tothe major common polypeptides for both wild-type and 'slime' sam-

ples. Gels were silver-stained.

50

Fig. 4. Kinetics of particle sedimentation after centrifugation at83000 x g for 3, 6, 9, and 28 h in 10-1l5% sucrose gradients ofcell-free extracts (1000 x g supernatants) from the wild-type strain ofN. crassa. White arrows indicate the position of the chitosomal chitinsynthetase peaks shown in Fig. 5 with respect to the turbidity bands

observed, which are identified by letters.

Kinetics of sedimentation of chitosomes us. other struc­tures in cell-free extract from wild-type

In order to ascertain whether or not the higherbuoyant density of wild-type chitosomes was caused byassociation with other subcellular particles responsible

for the overlapping UV-absorption (Fig. lA) andturbidity (Fig. 3, d) bands, we centrifuged a wild-typecell free extract (1000 X g supernatant) at 83 000 X g(R av ) for 3, 6, 9, and 28 h. Chitosomes sedimentedindependently of the major turbidity bands (arrows inFig. 4). After 3 h centrifugation, the chitin synthetasepeak was in a clear zone of the gradient (d ==1.0602-1.0666) separated from the faster movingturbidity bands; however, after 9 h, the chitosome bandhad overlapped with the uppermost turbidity band (d ==1.1187). Note that the cellular components responsiblefor the two major turbidity bands reached their iso­pycnic equilibrium after only 3 h centrifugation at83000 X g; prolonged centrifugation (6 and 9 h respec­tively) did not change the position of the turbiditybands (Fig. 4). The only significant difference was thatat 28 h centrifugation, the thin sharp band common toboth 'slime' and wild-type turbidity band patterns (Fig.3, c, g) had formed between the two diffuse wideturbidity bands (Fig. 4, b-d). This banding pattern alsoappeared by 19 h centrifugation (Fig. 3).

Fig. 5 shows that the chitosome peak and the majorUV-absorption bands from wild-type moved at differ-

I O<l 1 08 I 12 I 16 I 20 I 2<l

SPECIFIC GRAVITY

Fig. 5. Kinetics of sedimentation of chitin synthetase activity vs,major UV-absorbing structures. Samples (0.5 mil from II 1000 X gsupernatant (total chitin synthetase activity loaded, 3.45 units; specificactivity, 0.52 unitsy'mg of protein) were applied on top of 10-65%linear sucrose gradients. and centrifuged at 83000x g (R.y ) for 3, Il.9, and 28 h. Gradients were fractionated into 0.5 ml fractions, and 30III assayed for chitin synthetase. A z80 values correspond to undilutedsamples. Letters indicate samples taken for polypeptide composition

analysis (Fig. Il).

28h9h6h3h

particulate matter in the wild-type cell free extractbanded at the middle of the gradient in two broad,nearly confluent bands, separated by a sharp one (Fig.3, b-d), although a faint narrow band was also ob­served in the upper third of the tube (Fig. 3, a). Theseparation of the two heavy bands (Fig. 3, b, d) wasapparent even after 3 h of centrifugation at 83000 X g,and prolonged centrifugation (28 h) did not changetheir position in the gradients (Fig. 4, b, d). For 'slime'there were also four well-defined turbidity bands, twobroad ones in the upper third of the gradient (Fig. 3, e,f), a thin sharp band in the middle, that was common to'slime' and wild-type turbidity patterns (Fig. 3, g), andanother broad band slightly below this latter band (Fig.3, h).

The arrows in Fig. 3 point to the position of chitinsynthetase peaks shown in Fig. lA and lB, relative tothe turbidity bands for both 'slime' and wild-type.'Slime' chitosomes sedimented in a clear zone of thegradient just above the sharp middle band, whereas thechitosomes of wild-type co-sedimented with one of thetwo main turbidity bands (Fig. 3, d). This co-sedimenta­tion was not apparent in the electron microscope plates(Fig. lA, inset) because the negatively stained speci­mens we prepared probably excluded, or did not retain,the large subcellular fragments responsible for most ofthe turbidity.

51

Discussion

lated specific cell structures with the polypeptide bands,the overall banding pattern served as an extra marker tocompare the relative mobility of UV-absorbing struc­tures and chitosomes. During the first 19 h of centrifu­gation, fractions with. chitin synthetase activity dis­played the same set of polypeptides distributed in anM, range from 14000 to 55000, but the relative inten­sity of the bands for different fractions within eachpeak did not correlate with chitin synthetase activity(Fig. 6, lanes a- s). Furthermore, at 3, 6, 9 or 19 hcentrifugation, the most intense polypeptide bands werein fractions sedimenting above the chitosome peak frac­tion (Fig. 6, lanes a-s), whereas at the end of thecentrifugation (28 h), the banding pattern was invertedwith the most intensely stained bands below the chito­somal peak fraction (Fig. 6, lanes t-y). Therefore, theobserved polypeptide patterns correspond not to chito­somal proteins but to other more abundant structurespresent in the same fractions.

In agreement with previous results [9], we found thatdirect isopycnic centrifugation of crude cell-free extracts(1000 X g supernatants) on linear sucrose density gradi­ents, made in low molarity sodium phosphate buffer, ata slightly alkaline pH (8.2) to preserve chitin synthetaseactivity [19], was the most effective procedure to sep­arate functional chitosomes from N. crassa wild-typeand 'slime' variant strains. Under these conditions,where the risks of aggregation of subcellular organellesare minimized, we found a marked and consistent dif­ference in the buoyant density between wild-type and'slime' chitosomes. The discrepancy is not likely to bean artifact of cosedimentation with other cell particlessince we showed that the various types of subcellularcomponents detected by UV-absorption, turbidity, orchitin synthetase assay sediment independently of oneanother, although some reach the same equilibriumdensity after prolonged centrifugation.

The value of buoyant density (d = 1.125) previouslyreported [9] for the chitosomes of 'slime' falls within therange of median values measured here (d = 1.1201 ±0.0036). This seems an unusually low value, not onlylower than the value for chitosomes from the myceliumof N. crassa (d = 1.1349 ± 0.0024), but also lower thanthe values reported for chitosomes from the yeast andmycelial forms of M. rouxii [12].

'Slime' cells seem to have all the cellular componentsnecessary for chitin synthesis [7-9], and they also syn­thesize and release into the medium cell-waIl-related(periplasmic) enzymes such as invertase and acid phos­phatase [20-23], and, possibly, some self-assemblingprotein components of the wild-type cell walls (Martinez,Gil, Casanova, Rico, Sentandreu and Ruiz-Herrera,submitted for publication), yet they fail to make a

-,~

- 1 ~

- 66

- 45

- 31

- :11

- 92

nopqrs

- - -- - . ..---- ..... - -- ~--

--- -....... ------

28h .

-1 ~ ..9 h k m '0

T"

9h.... ... - -

)(

- 92 s~

- 66

3h - r:-

Fig. 6. Polypeptide composition detected in the lettered fractions fromFig. 5, following the experimental protocol described in Fig. 2. Silver

staining.

- 92

-66

•• ;.;;;.. ::: - ~S----_ - 14

t uvw x y

ent rates but by 28 h all had merged into one broadpeak; in fact, after 19 h centrifugation (Fig. 1A) theywere already merged. The chitosome peak attained itsbest separation from the major bands of UV-absorptionfor a brief period, after about 6-9 h of centrifugation(Fig. 5).

The polypeptide composition of fractions from thechitosomal peaks was analyzed by SDS-PAGE as anadditional criterion to follow the distribution patternsof subcellular particles. Although we have not corre-

52

normal cell wall. This inability may result from a com­bination of diverse biosynthetic deficiencies. Thus,according to Leal-Morales and Ruiz-Herrera [10], lackof detectable gluean synthetase activity in 'slime' couldbe one cause of the inability to synthesize a cell wall.Our finding that chitosomes from 'slime' variant havelower buoyant density than those from the wild-typeoffers a plausible lead to account for the inability tomake cell wall chitin. 'Slime' chitosomes, though oper­ational in vitro, may have specific alterations in struc­ture and/or composition that make them unable tooperate in vivo. The lower buoyant density of 'slime'chitosomes suggests the existence of differences in lipidcontent and/or composition rather than protein com­position; changes in protein composition apparently donot alter sedimentation properties of other types ofvesicles, e.g. plasma membrane vesicles [19]. Since thechitosomal chitin synthetase from 'slime' is fully func­tional in vitro [9] we think. that any such compositionaldifferences would affect not the catalytic properties ofchitin synthetase, but rather the interaction betweenchitosomes and other structures needed for chitosomesto migrate to, and interact with, the plasma membranein vivo. The conjectured defective interaction betweenchi tosomes and plasma membrane might be due to thechitosomes, since no differences have been found inchemical properties, functional behavior, or proteincomposition between the plasma membranes of •slime'and wild-type N. crassa [20-23].

Acknowledgements

This work was supported in part by a grant (GM­33513) from the National Institutes of Health, U.S.A.J.P.M. was a recipient of Postdoctoral Fellowships fromthe Tratado de Amistad y Cooperacion entre Espana ylos Estados Unidos de America (Acuerdo Complemen­tario No.3), and from the Fondo de InvestigacionesSanitarias de La Seguridad Social, Ministerio de Sani­dad y Con sumo (Spain).

References

1 Bartnicki-Garcia, S., Ruiz-Herrera, J. and Bracker, C.E. (1979) in:Fungal Walls and Hyphal Growih (Burnett, J.H. and Trinci,

A.P.I., eds.), pp. 149-168, Cambridge University Press, Cam­bridge, U.K.

2 Bracker, C.E., Ruiz-Herrera, J. and Bartnicki-Garcia, S. (1976)Proc. Natl. Acad, Sci. USA 73, 4570-4574.

3 Hanseler, E., Nyhlen, L.E. and Rast, D.M. (1983) Exp. Mycol. 7,17-30.

4 Ruiz-Herrera, 1., Lopez-Romero, E. and Bartnicki-Garcia, S. (1977)J. BioI. Chern. 252, 3338-3343.

5 Bartnicki-Garcia, S., Bracker, C.E., Reyes, E. and Ruiz-Herrera, J.(1978) Exp. Mycol. 2,173-192.

6 Emerson, S. (1963) Genetica 34,162-182.7 Wiltse, J.A. (1969) Cell wall synthesizing enzymes in a mutant of

Neurospora crassa lacking cell walls. Ph.D. Dissertation, Univer­sity of Minnesota, Dissertation Abstracts No. 70-15, 837.

8 Selitrennikoff, CiP, (1979) Biochim. Biophys. Acta 571, 224-232.9 Bartnicki-Garcia, S., Bracker, C.E., Lippman, E. and Ruiz-Herrera,

J. (1984) Arch. Microbiol. 139,105-112.10 Leal-Morales, C.A. and Ruiz-Herrera, J. (1985) Exp. Mycol. 9,

28-38.11 Martinez, J.P., Gimenez, G. and Bartnicki-Garcia, S. (1987) Exp.

Mycol. 11, 278-286.12 Ruiz-Herrera, J., Bracker, C.E. and Bartnicki-Garcia, S. (1984)

Protoplasma 122,178-190.13 Vogel, H.J. (1964) Am. Nat. 98, 435-466.14 Beadle, G.W. and Tatum, E.L. (1945) Am. J. Bot. 32, 678-686.15 Scarborough, G.A. (1975) J. BioI. Chern. 205, 1106-1111.16 Ruiz-Herrera, J. and Bartnicki-Garcia, S. (1976) J. Gen. Microbial.

97, 241-249.17 Laemmli, U.K. (1970) Nature 227, 680-685.18 Wray, W., Boulikas, T., Wray, v.P. and Hancock, R. (1981) Anal.

Biochem. 118, 197-203.19 Arroyo-Begovich, A. and Ruiz-Herrera, 1. (1979) J. Gen. Micro­

bial. 113, 339-345.20 Metzenberg, R.L. (1963) Biochim. Biophys. Acta 77,455-465.21 Bigger, C.H., White, M.R. and Braymer, H.D. (1972) J. Gen.

Microbial. 71, 159-166.22 Casanova, M., Martinez, J.P., Gil, M.L., Sentandreu, R. and

Ruiz-Herrera, J. (1987) 1. Gen. Microbial. 133, 2447-2456.23 Ruiz-Herrera, J., Martinez, J.P., Casanova, M., Gil, M.L. and

Sentandreu, R. (1987) Arch. Microbiol. 149, 156-162.24 Das, a.p. and Henderson, E.J. (1983) Biochim. Biophys. Acta 736,

45-56.25 Brooks, K.M., Addison, R. and Scarborough, G.E. (1983) 1. BioI.

Chern. 258, 13909-13918.26 Scarborough, G.A. (1978) Methods Cell Bioi. 20,117-133.27 Stotish, R.L. and Samberg, E.W. (1981) Biochim. Biophys. Acta

641, 289-300.28 Stroobant, P., Dame, J.B. and Scarborough, G.A. (1980) Fed.

Proc. 39, 2437-2441.