Assessment of Elasticity and Topography of Aspergillus nidulans Spores via Atomic Force Microscopy

Previous studies have described both surface morphology andadhesive properties of fungal spores, but little informationis currently available on their mechanical properties. In thisstudy, atomic force microscopy (AFM) was used to investigateboth surface topography and micromechanical properties of Aspergillusnidulans spores. To assess the influence of proteins coveringthe spore surface, wild-type spores were compared with sporesfrom isogenic rodA⁺ and rodA^– strains. Tapping-mode AFMimages of wild-type and rodA⁺ spores in air showed characteristic"rodlet" protein structures covering a granular spore surface.In comparison, rodA^– spores were rodlet free but showeda granular surface structure similar to that of the wild-typeand rodA⁺ spores. Rodlets were removed from rodA⁺ spores bysonication, uncovering the underlying granular layer. Both rodlet-coveredand rodlet-free spores were subjected to nanoindentation measurements,conducted in air, which showed the stiffnesses to be 110 ±10, 120 ± 10, and 300 ± 20 N/m and the elasticmoduli to be 6.6 ± 0.4, 7.0 ± 0.7, and 22 ±2 GPa for wild-type, rodA⁺ and rodA^– spores, respectively.These results imply the rodlet layer is significantly softerthan the underlying portion of the cell wall.

INTRODUCTION

Top

Introduction

Fungal spores are of importance medically (25), agriculturally(9), and industrially (3) and provide fungi a natural meansof reproduction, dispersion, and survival in adverse environmentalconditions (22, 29). These benefits are in large part due tothe spore cell wall, which has been compared to a man-made compositematerial (24). The wall is typically composed of various polysaccharideand protein components (17) and imparts great strength and resistanceto chemical attack (24).

Knowledge of spore cell wall mechanical properties is necessaryfor a complete understanding of molecular component ultrastructure.Spore mechanical properties are also relevant during wall expansionwhich occurs as part of the germination process. It has beenpointed out that changes in cell wall mechanical propertiesare central to the emergence of the germ tube (6, 21). Whilea number of studies have described surface morphology (4, 13,14, 31) and adhesive properties (7, 27) of fungal spores, noinformation is currently available on their relevant micromechanicalproperties (e.g., elasticity).

The elasticity of an object can be described in terms of stressand strain. Stress is defined as the force applied per unitarea, while strain is the resulting amount of deformation perunit length. The ratio of stress to strain (for an elastic materialfollowing Hooke's law) is defined as the elastic modulus (E),and describes the mechanical resistance of a material duringelongation or compression. A large E implies a stiff or strongmaterial, while a small E implies a softer material. To measureE, as well as other micromechanical properties, of differenttypes of biological materials, a number of authors have usedatomic force microscopy (AFM) (1, 2, 32, 33). To carry out thesetypes of tests, a rigid AFM probe is used as an "indenter" ofthe soft biological material, and the generated force-displacementdata, or "force curves," are then used to calculate the elasticmodulus of the sample (28).

Previous electron microscopy studies have shown "rodlet" structureson the Aspergillus nidulans spore wall (18). These are composedprimarily of protein (4, 10, 31) and are thought to serve severaldifferent functions (31). Rodlet structures are related to thewall structural gene rodA (26), in that rodA^– spores lackthe rodlet layer and are less hydrophobic than rodA⁺ spores(26). The goal in this work was to use AFM to study spores ofA. nidulans by investigating surface morphology and calculatingtheir elastic modulus. Three different A. nidulans strains (i.e.,wild type, rodA⁺, and rodA^–) were studied and compared.Surface morphology of spores was imaged by tapping-mode AFM,and force displacement measurements were used to determine theelastic modulus.

MATERIALS AND METHODS

Top

Materials and Methods

Sample preparation.
All A. nidulans strains used in this study were obtained fromthe Fungal Genetics Stock Center (Kansas City, Kans.): Glasgowwild type (A4), rodA^– (A849; pabaA1 yA2 ${Delta}$ argB::trpC ${Delta}$ B ${Delta}$ rodA::argBveA1 trpC801), and rodA⁺ (A851; pabaA1 yA2 ${Delta}$ argB::trpC ${Delta}$ B veA1trpC801). Strains were stored as a frozen stock culture (5)and grown on potato dextrose agar (Difco, Detroit, Mich.) platesat 32°C for 3 to 4 days for sporulation. Round (12-mm diameter)glass coverslips (Fisher Scientific, Pittsburgh, Pa.) were cleanedwith ethanol, coated with 0.01% poly-L-lysine (Sigma, St. Louis,Mo.) for 5 min, and used as the substrate for spore samplesproduced by the following two methods. "Untreated" spores werescratched from sporulated mycelial mats with an inoculatingloop and, without further treatment, tapped over a poly-L-lysine-coatedcoverslip which was then used for AFM. Alternatively, "sonicated"spores were obtained from agar plates, suspended in steriledeionized water, and subjected to sonication (550 Sonic Dismembrator;Fisher Scientific, Pittsburgh, Pa.) at 4°C for 10 min witha 20% power output. After rinsing twice in deionized water,a drop of spore suspension was pipetted onto a poly-L-lysine-coatedcoverslip, allowed to dry in air (approximately 30 min), andsubjected to AFM testing.

AFM imaging and force measurements.
All experiments were performed in air, using a multimode atomicforce microscope (Nanoscope IIIa; Digital Instruments, SantaBarbara, Calif.) equipped with a J-type piezoscanner. Both rotatedTappingMode Etched Silicon Probes (RTESP; Digital Instruments)and Olympus TappingMode Etched Silicon Probes (OTESP; DigitalInstruments) were used for the imaging. Amplitude and heightimages were obtained in the tapping mode with a scan rate of1 Hz and an integral gain of 0.3 to 0.5. The tapping force wasadjusted by changing the set point voltage until high-resolutionimages were obtained in minimal tapping force. All images wererecorded at room temperature and approximately 50% humidity.To perform force measurements, a spore was scanned in tappingmode to obtain a high-magnification image and to locate a positionon the spore for force measurements. The cantilever tip wasthen withdrawn, and a force displacement curve was taken usingthe "trigger mode" (i.e., the piezo rises vertically until thepreset maximal cantilever deflection, or maximal applied force,is reached and then retracts a distance equal to the presetvertical scan size). The cantilever deflection was calibratedby taking force curves on bare coverslips. To avoid large variationof spring constants of individual cantilevers, only one RTESPcantilever and one OTESP were used in all force measurements.The spring constants of these cantilevers were determined tobe 61 and 82 N/m, respectively, by measuring the resonance frequenciesof the cantilevers (11). The silicon tip was found to have aradius of 10 nm by scanning several calibration gratings andextracting the tip shape from the resulting image (data notshown). Force curve data were used to calculate stiffness andE as described in the next section. Statistical comparison wasperformed by using single-factor analysis of variance, and theresults are presented with the P value.

Theory.
The presented micromechanical calculations assume that, in responseto the external, concentrated, normal force exerted by the AFMtip, the A. nidulans spore wall is indented instead of bentor stretched. This assumption is based on the fact that A. nidulansspores possess cell walls several hundred nanometers in thickness(6), while the deformation depth involved in this study is only5 to $~$ 10 nm.

Pharr et al. (23) showed that for any axis-symmetric indenterswith smooth profiles, the unloading stiffness, dF/d ${delta}$ , is relatedto the projected contact area, A, and the reduced elastic modulus,E_r, in the following equation:

(1)
HereF is the applied force and ${delta}$ is the indentation depth.

In the contact region of the force curves, the displacementof the sample will produce a deflection of the cantilever. Ifboth the cantilever and the sample are infinitely hard, thenthe cantilever deflection, d, will equal the sample displacement,z (assuming the initial contact point as the origin). Otherwise,the sample displacement equals the sum of the cantilever deflectionand the indentation on the specimen:

(2)
Theapplied force, F, can be determined by assuming Hooke's lawbehavior (valid for small displacements):

(3)
wherek is the cantilever spring constant, determined by measuringresonant frequency and using known geometric and material propertiesof the cantilever (11):

(4)
where l,w, ${rho}$ , E_i, and ${nu}$ _o are the length, width, density, E, and resonancefrequency of the cantilever, respectively.

At small indentation depths, as in this work, the apex of theAFM tip can be assumed to have a spherical profile with a radiusof curvature, R. This assumption was tested by imaging the AFMprobes used here over sharp calibration spikes. It was foundthat the contour of the lower portion of the tip could be approximatedwell by a hemisphere (data not shown). The projected area ofelastic contact is then given by the geometry (Fig. 1A)

(5)
where ${delta}$ _p is the indentation depth below the circleof contact which is found from reference 15

(6)
andthe maximal indentation depth, ${delta}$ _t, and residual depth, ${delta}$ _r, aredetermined experimentally (Fig. 1B).

View larger version (14K):
[in this window]
[in a new window]
FIG. 1. Schematic illustration of (A) elastic-plastic contact between a rigid spherical indenter and a specimen and (B) concurrent load versus displacement curve used to determine stiffness and the elastic modulus. R is the indenter radius, F_t is the maximum applied force, ${delta}$ _p is the indentation depth below the circle of contact, ${delta}$ _t is the maximum indentation depth, ${delta}$ _r is the residual depth, and dF/d ${delta}$ is the unloading stiffness.

The initial portion of the unloading curve represents an elasticcontact, and the slope dF/d ${delta}$ is defined as the stiffness, S (Fig.1B), and calculated from the observed slope, m, of the experimentalforce curves (d versus z) according to

(7)
Inequation 1, E_r is used to correct the effects of nonrigid indenterson penetration measurement. Thus, the elastic modulus of thematerial can be calculated according to the following relation(23):

(8)
where E and E_i are the elasticmoduli and v and v_i are the Poisson ratios of the specimen andthe indenter, respectively. Typically, soft biological tissuesare treated as incompressible materials (Poisson ratio = 0.5)because of their high water content. However, in this study,testing was conducted in air and thus water content of sporeswas likely to be relatively low (21). When this is the case,one can generally expect values of the Poisson ratio in theinterval from 0 to 0.5 for most practical materials (8). Wenote that in this range variations for corresponding E valuesdetermined from equation 8 are less than 15% of the modulusfor the maximum v (= 0.5); therefore, an average value v = 0.3was assumed for the spores. For a silicon indenter, E_i and v_iare 130 GPa and 0.28, respectively (http://www.ioffe.rssi.ru).

RESULTS AND DISCUSSION

Top

Results and Discussion

Spore surface morphology.
Tapping-mode AFM was used to assess fungal spore surface morphology(Fig. 2). Untreated wild-type spores (Fig. 2A) had a rough surface,consisting of granular domains (approximately 0.4 µm inlength) with a rodlet layer on each. By comparison, rodA⁺ spores(Fig. 2B) showed similar rodlet structures on smaller domains(approximately 0.2 µm in length), while rodA^– spores(Fig. 2C) exhibited granular materials on a rodlet-free surface.Previous studies have shown that rodlet structures are directlyrelated to the rodlet-encoding gene rodA and that deletion ofrodA leads to loss of the rodlet layer (26). Our AFM imagesare consistent with these findings. Additionally, the observationthat rodA⁺ spores exhibit smaller rodlet domains than wild-typespores indicates that the rodA⁺ strain may harbor a mutationaffecting rodlet domain organization.

View larger version (139K):
[in this window]
[in a new window]
FIG. 2. Images generated by AFM in the tapping mode of the surface morphology of (A to C) untreated and (D to F) sonicated spores. (A and D) wild type; (B and E) rodA⁺; and (C and F) rodA^–. Bar length, 0.2 µm.

To determine the affinity of the rodlet layer for the sporesurface, A. nidulans spores were subjected to AFM imaging aftersonication (Fig. 2D to F). Both wild-type and rodA⁺ strainswere observed to lose nearly all rodlets, uncovering the underlyingregion of granular materials. In comparison, no perceivablemorphological changes were observed on rodA^– spores beforeand after the sonication (Fig. 2C and F). Furthermore, we foundthe surface features of the sonicated rodA⁺ spores were quitesimilar to those of rodA^– spores. These observations implythe rodlet layer does not adhere strongly to spore surface andcan be removed by hydrodynamic shear.

Higher-resolution images revealed additional rodlet detail (Fig.3). On both wild-type and rodA⁺ strains, the rodlet layer coveredthe entire surface of each domain, with each rodlet being approximately10 nm in diameter and hundreds of nanometers in length. Theseobservations are consistent with previous studies employingelectron microscopy (14, 16) or contact mode AFM (13). Our images,however, reveal additional detail, as each rodlet appears tobe composed of two strands, each of which is approximately 3nm in diameter (Fig. 3D). Previous studies have shown rodletsare composed primarily of the protein hydrophobin (20, 30, 31),whose diameter is speculated to be approximately 3.1 nm (12),in close agreement with the strand diameters found here.

View larger version (98K):
[in this window]
[in a new window]
FIG. 3. Images (A; height) and (B; amplitude) show the rodlet structure on the wild-type spores. (C) Cross-section along the line in panel A reveals the periodicity of the rodlets. Image D shows rodlets are apparently composed of two strands. Bar lengths: 50 nm in panels A and B and 10 nm in panel D.

Apparent elastic moduli of fungal spores.
With the cantilever tip used as a microindenter, force-displacementmeasurements (i.e., force curves) were made and used to calculatethe elastic modulus of the spores. As the normal spore wallapparently has at least two layers (rodlet and underlying granularmaterial), with their true elastic moduli unknown, an apparentelastic modulus, Ê, was defined to represent responsesfrom both layers in force displacement measurements.

To assess reproducibility, force curves were repeatedly takenat the same position on a number of wild-type spores (Fig. 4).For the first several cycles, the approaching curves are notreproducible, implying plastic deformation or viscoelasticitythat is not reversible on the time scale of the measurement.After enough cycles, force curves are reproducible, but insteadof overlapping, the loading and unloading force curves exhibithysteresis, implying viscous contributions from the spore. Becausethe initial portion of the unloading curves represents an elasticcontact, data from unloading curves were used for calculationof Ê.

View larger version (25K):
[in this window]
[in a new window]
FIG. 4. Typical force curves for five sequential tests at the same position on the same untreated wild-type spore, measured with an RTESP (k = 61 N/m). Loading curves (thick lines) are labeled 1 through 5. Unloading curves (thin lines) overlap.

To determine the effect of the maximum force applied by theAFM cantilever (i.e., trigger threshold deflection), a seriesof force curves were collected at the same position on a wild-typespore, increasing the maximum applied force in each subsequenttest. Small indentations (normally several nanometers) wereapplied. As shown in Fig. 5A, the depth of cantilever indentationincreased with the maximal force applied (a visible permanentimpression was made on the spore surface at 2,400 nN; data notshown). Ê calculated from these tests also increased correspondingly(Fig. 5B), implying that the underlying layer was more rigidthan the rodlet layer as the tip sensed more and more contributionsfrom the underlying layer. Thus, to ensure consistent measurementsof Ê, force curves were generated with a maximum appliedforce of 650 nN (unless specified).

View larger version (9K):
[in this window]
[in a new window]
FIG. 5. Effect of increasing maximum applied force on (A) indentation depth and (B) apparent elastic modulus, Ê, for a wild-type spore, measured with an RTESP (k = 61 N/m). Error bars represent the standard error of the measurements shown in Fig. 7.

Typical force curves taken both on bare glass (calibration)and of different spore samples are shown in Fig. 6. One notabledifference between the four force curves is that they have differentslopes in the contact region. Glass is assumed to be infinitelyhard and has a steep linear contact region (slope = 1; Fig.6A). In the spore force curves (Fig. 6B to D) the lower slopeimplies a softer surface. Force curves were taken on a numberof different spore samples (5 to 11 spores for each category,3 to 5 force curves for each spore at different positions);values of stiffness (S) were determined to be 110 ± 10,120 ± 10, and 300 ± 20 N/m; and apparent elasticmoduli (Ê) were determined to be 6.6 ± 0.4, 7.0± 0.7, and 22 ± 2 GPa for untreated wild-type,rodA⁺, and rodA^– spores, respectively (Fig. 7). The tworodA-bearing strains (e.g., wild type and rodA⁺) had similarS (P = 0.15) and Ê (P = 0.57) values. In contrast therodA^– strains had larger S and Ê values than therodA⁺ strain (P < 0.001). This implies the rodlet-coveredsurfaces are softer than the rodlet-free surface. However, thisdifference was eliminated after the removal of the rodlet layer.When both rodA-bearing strains were sonicated, they exhibitedsimilar granular, rodlet-free surfaces and had similar S (P= 0.62) and Ê (P = 0.11) values (Fig. 7). Typically, sonicatedspores had larger S and Ê values than untreated spores,partially due to the loss of the rodlet layer. In addition,it is noted that rodA^– spores also showed slightly largerS and Ê values after the sonication (without a perceivablemorphological variation; see Fig. 2C and 4C), which impliesthe sonication or other unknown factors affect the determinationsof mechanical properties.

View larger version (16K):
[in this window]
[in a new window]
FIG. 6. Force-displacement curves measured on the surface of (A) bare glass, (B) rodA^–, (C) rodA⁺, and (D) "sonicated" rodA⁺ with an OTESP silicon probe, k = 82 N/m. Dashed lines in the contact region represent initial slopes (m) of the unloading curves, which were determined to be 0.79, 0.60, and 0.84, respectively, for rodA^–, rodA⁺, and "sonicated" rodA⁺ spores.

View larger version (18K):
[in this window]
[in a new window]
FIG. 7. Average values of (A) the apparent elastic modulus, Ê, and (B) the stiffness, S, for the wild-type, rodA⁺, and rodA^– A. nidulans spores, measured with an RTESP (k = 61 N/m on the wild type) and an OTESP (k = 82 N/m on rodA⁺ and rodA^– spores). Error bars represent the standard error of the measurements.

It has been shown that fungal cell walls are sensitive to KOH,laminarinase, pronase, and chitinase treatment (19). Accordingto these results, it was suggested that wall components suchas amorphous ${alpha}$ -1,3 glucan, ß-1,3 glucan, and glycoproteinsmake up the wall matrix, while cross-linked polysaccharidessuch as chitin are primarily maintained in the inner wall layers(19). Our data imply that the rodlet-covered surface is softerthan the rodlet-free surface. Thus, the rodlet protein surfacelayer is apparently softer than the complex matrix of cross-linkedpolysaccharides in the underlying wall structure. This matriximparts great strength and has been compared to a man-made compositematerial (24). The relatively high wall elastic modulus we determinedimplies it is difficult for dormant spores to germinate, inwhich the yielding of the cell wall to the increasing turgorinside the cell leads to cell germination and the protrusionof germ tubes. Therefore, a decrease in wall strength or wallsoftening is assumed to accompany spore germination (21).

Conclusion.
While a number of studies are available on spore surface morphology,little is know about spore micromechanical properties. In thisstudy, AFM was used to visualize the surface of fungal sporesand to make measurements of the spore wall stiffness and elasticmodulus. In agreement with others, we find the spore surfaceto be covered with a rodlet layer, apparently composed of protein.This layer was easily removed by sonication. Tests of sporeswith and without this rodlet layer show the stiffness and elasticmodulus of rodA⁺ spores are approximately one-third the valuesof rodA^– spores. This implies that the rodlet layer issignificantly softer than the underlying portion of the cellwall.

ACKNOWLEDGMENTS

This work was supported by grants BES-9876012 and BES-9906586from the National Science Foundation and by Novozymes NorthAmerica, Inc.

We thank M. P. Nandakumar for his assistance in fungal culture.

FOOTNOTES

* Corresponding author. Mailing address: Department of Chemical and Biochemical Engineering, University of Maryland, Baltimore County (UMBC), 1000 Hilltop Circle, Baltimore, MD 21250. Phone: (410) 455-3439. Fax: (410) 455-1049. E-mail: marten{at}umbc.edu.

REFERENCES

Top