PLANT PHYSIOLOGY

SummaryAn improved spectrophotometric method for the

determination of lipoxidase activity was developedand applied in studies of the purified enzyme andcrude enzyme preparations from leguminous seeds,with linoleic acid solubilized in Tween 20 as thesubstrate. The optimum pH was found to be 7.0, 6.5,6.0, and 5.5, for purified soybean lipoxidase and forthe crude lipoxidases extracted from gram flour (Ci-cer sp.) soybean meal (Glycibic miax L.), and hy-(lrate(l mung beans (Phaseolhs auircuEs L.), respec-tively. The application of this test illustrates A) thatthe present method was free from the inherent limita-tions on pH present in the original methods, B) thatincreasing the amount of detergent, in a fixed concen-tration of fatty substrate, caused inhibition in theenzymatic activity; more pronounced inhibitionsoccurred when the concentrations of detergent as wellas of fatty substrate were increased in equal propor-tions, and C) peroxide formation is proportional totime of reaction and to enzyme concentration.

Determination of lipoxidase activity in germinat-ing mung beans indicated 3 periods of enzyme activ-ity: a slow decline from 8 to 24 hours, a rapid declinefrom 24 to 40 hours, and a linear decline from 40 to80 hours of germination.

Literature Cited1. BERGSTR6M, S. 1946. On the oxidation of linoleic

acid with lipoxidase. Arkiv Kemi NMIinieral Geol.Band 21 A. No. 15.

2. FRITZ, G. J. 1962. Studies on 02-fixation by maizeseedlings. Plant Physiol. 37: xxxii-xxxiii.

3. HOLMAN, R. T. AND S. BERGSTR6dM. 1951. Lipoxi-dase or unsaturated-fat oxidase. Inl: The EnzymeChemistry and -Mechanism of Action, J. B. Sumnerand K. NMyrbkck, eds. Academic Pres ILnc., NewYork. vol. 2, Part 1, 559-74.

4. HOILMAN, R. T. AND G. 0. BlURR. 1945. Spectro-photometric studies of the oxidation of fats. IV.Ultraviolet absorption spectra of lipoxidase-oxidizedfats. Arch. Biochem. 7: 47-54.

5. HOLMAN, R. T. 1948. Lipoxidase activity andfat composition of germinating soy beans. Arch.Biochem. 17: 459-66.

6. HOLMAN, R. T. 1955. Measurement of LipoxidaseActivity. In: Methods of Biochemical Analysis,D. Glick. ed. Interscience Publishers, DNew York.vol II, 113-19.

7. TAPPEL, A. L. 1962. Lipoxidase. In: Methods inEnzymology, S. P. Colowick and N. 0. Kaplan,eds. Academic Press., New York anid London.vol 5, 539-42.

8. THEOREILI, H., S. BERGSTR6M, AND A. AKESON. 1946.Activity determination and further purification oflipoxidase. Pharm. Acta Helv. 21: 318-24.

Chemotropic Response of the Pollen of Antirrhinum majus to Calcium ,Joseph P. Mascarenhas3 and Leonard MachlisDepartment of Botany, University of California, Berkeley 4

We reported earlier (A) our failure to identifyamong a large number of organic compounds activein biological systems any that exerted a chemnotropiceffect on the pollen tubes of snapdragon and (B) theresults of extraction and fractionation procedures thatsuggested the chemotropic factor from gynoecia tobe a quite small molecule, heat stable, water soluble,and associated with larger molecules from which itcould be separated by various means (14). This

'Received May 6, 1963.2 Most of the research reported in this paper is part of

a dissertation submitted by the first author to the GraduateDivision of the University of California at Berkeley inpartial satisfaction of the requirements for the degreeof Doctor of Philosophy in the field of Plant Physiology.The work was supported by research grant G-7031 fromthe National Science Foundation.

3Present address: Department of Biology, WellesleyCollege, Mass.

led to a search among inorganic ions, l)lartictilarlythose reported at one time or another to enhancegermination andl tube growth of pollen. \When wetested the chlorides of Ca, AIn, and Zn at severaldifferent concentrations, AMn and Znl prove(l to beinactive but Ca elicited a pronounced chemotropicresponse. The subsequent investigation of the chemio-tropic effect of Ca is reported in this paper. A briefsummary of a part of these studies N\\aslpreviouslypublished (15).

Materials and MethodsThe basic procedures and materials are (Iescribed

in detail in the earlier report (14). Flowers of An-tirrhinuw itajuts (tetraploid) grown in the UniversityBotanical Garden provided both gynoecia an(l pollen.For a brief period wlhen these plants were not avail-able, cut flowers were purchased from a local florist.

70

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

MASCARENHAS AND MACHLIS-CHEMOTROPIC RESPONSE

OVULES POLLEN PAPER

FIG. 1. Schematic diagram of the arrangement ofthe depressions in agar and the positioning of materialsin the depressions for the assay of chemotropism.

The apparatus for the assay of chemotropic activ-ity was an agar plate in which were formed 15 setsof 3 depressions as diagrammed in figure 1. Thefloor of the center (pollen) chamber was coated withpollen by means of a fine brush. The material to betested was placed in one of the end (test) chambersagainst the wall facing the pollen chamber. Thematerials tested were of various forms. Solid mate-rials such as plant parts, CaCO3, and calcium malatewere placed directly in the test chamber. Solutionswere tested in either of 2 ways. Infrequently, agarwas added to a solution to give a concentration of 1 %.While still melted, a drop equal in volume to the testchamber was added to the chamber where it rapidlysolidified. More often, measured amounts of solu-tions were absorbed into a 3 X 5 mm piece of What-man No. 1 filter paper which was then dried withgentle heat and positioned in the test chamber againstthe dividing wall. The third or control chamberusually was left empty, particularly after it was foundthat matching the content of this chamber with thatof the test chamber, except for the substance undertest, gave results no different than when the controlchamber was left empty. After the apparatus wasassembled, it was incubated for 8 hours at 250 in thedark. Readings entailed counting the number (andfrequently measuring the length) of the pollen tubesthat grew into the agar walls separating the pollenchamber from the test and control chambers. Atypical response to ovules is shown in figure 2a. Thenumber of tubes growing into the test wall is pro-portional to the concentration of active material upto tube counts of approximately 200. The systemapparently saturates a little beyond this level; more-over, the tubes are so crowded that accurate count-ing becomes impossible.

The basic nutrient medium (hereafter referred toas the plate medium) used in the bioassay consistedof 10 or 20% sucrose (depending on the lot of pollen),1% yeast extract (Difco), 0.01%0 boric acid, and 1%,agar (Difco ) 4. The optimum concentrations of su-crose and yeast extract for germination and tubegrowth were determined by appropriate tests.

The concentration of calcium in tissues was deter-mined as follows. Tissues were dissected from flow-ers, dried in an oven at 850 for 24 hours, and weighed.The oven-dry material was then ashed in a mufflefurnace at 550° until free of carbon. An acid extractof the ash was prepared and after removal of inter-fering ions, calcium was determined by titration withcyclohexanediamine-tetraacetic acid using calcein asthe indicator (4).

Flowers containing Ca45 were obtained as fol-lows. A snapdragon plant growing in the BotanicalGarden was lifted from the ground on MIarch 8, 1962and the roots washed free of soil. The plant was thenplaced in a heated greenhouse in a jar containing 3.5liters of nutrient solution [see Complete (FeEDTA)in table I of Machlis and Torrey (13)]. The nutrient

Table IThe Effect of Calcium as (CaC12) onl the Directional

Growth of Pollen TubesThe plate medium consisted of 10% sucrose, 0.1%

yeast extract, 0.01% boric acid, and 1%7 agar. Nothingwas placed in the control chamber. The readings weremade after incubation for 8 hours at 25'. Each value isthe average of 4 replicates.

Growth to calcium Growth to controlCalciumagper Number Length Number Lengthpaper (mm) (mm)

0.0 21 0.3-0.4 20 0.3-0.40.8 46 0.3-0.5 23 0.3-0.41.6 70 0.3-0.6 27 0.3-0.42.4 90 0.3-0.6 26 0.3-0.44.0 110 0.3-0.6 26 0.3-0.46.0 198 0.4-0.8 33 0.3-0.58.0 226 0.5-0.9 45 0.3-0.59.6 260 0.5-0.9 47 0.3-0.6

20.0 269 0.540.9 127 0.34.6

solution was changed periodically until Mfay 14, 1962at which time the first floral buds were just beginningto appear. At this time the plant was transferred toa jar containing fresh medium but in which theCa(NO3)2 was replaced with 1.0 mc of Ca45C12.During the next several weeks flowers were collectedimmediately after the dehiscence of anthers. Theradioactivity of ashed samples was determined undera Micromil window tube in an atmosphere of Q gas.

4 To eliminate any possibility of ambiguity over theuse of concentrations expressed as per cent the procedurewith sucrose, yeast extract, boric acid, etc., was to takethe appropriate weight of solute and dilute (with solution)to the final volume, i.e., for 10% sucrose 1Og of sucrosewas diluted with water to 100 ml. With agar, for technicalreasons, the appropriate weight was added to the finalvolume of solution, i.e., for 1% agar 1 g of agar wasadded to 100 ml nutrient solution.

71

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

Results

The response of pollen tubes to calcium is pre-sented in table I. The amounts of calcium indicatedwere added to rectangles of filter paper from a meth-anol solution of 1 X 10-2 At CaCl2. The numberof tubes growing towards the source of calcium isproportional to the amount of calcium until satura-tion is reached. Although the higher numbers oftubes could not be counted accurately attempts to doso were made and indicated that there was no furthersubstantial increase in the number of tubes that grewinto the wall. The data in table I also show thatthe higher amounts of calcium cause a growth of tubestowards the control chamber in excess of the back-ground number. It is presumed that calcium diffusedacross the pollen chamber into the control wall whereit then caused tubes to grow towards it. If the assayis mia(le at the end of 3 hours instead of 8 hours, thenumber of tubes growing into the control wall is thesame at all concentrations of calcium.

The cation calcium anld not the anion associatedwith it is the cause of the response of the tubes; CaC12,Ca (NO,)2. CaSO4, calcium gluconate, and calciumcaseinate were all active. The chlorides of Mg, Ba.Sr, Na, ancl K over a wi(le concentration range(10 Aliter per paper rectangle of solutions of 1X 10-4, 1 X 10-3, 1 X 10-2, 1 X 10-1, and1.0 AI) failed to elicit a response, thus demonstratingthe specificity of calcium. In these tests pH was nota factor. Noninhibitory (of germination) solutionsof these salts (all except 1.0 aM\ MgCl2, 1 X 10-1 and1.0 :r BaCl, 1 X 10-2, 1 X 10-1, and 1.0AtSrCl,) had pH values that overlapped extensivelywith those of active solutions of CaC1, and Ca(NO3)2(pH 4.154-.30).

Very few pollen tubes grow out of the pollenchamber unless calcium is supplied from the testchamber. It will be proposed later that the tubesstay in the pollen chamber in response to calciumreleased from the pollen grains. What happens ifthe plate medium itself contains calcium? As might

be expected, when all of the agar surrounding thepollen chamber (bottom and all sides) contains suf-ficient calcium the tubes grow out of the pollen cham-ber in all directions. This response is shown intable II. Important parameters from these data tobe used later are that growth of tubes out of the pollenchamber in all directions becomes high with concen-trations of calcium in excess of 1 X 10 AI, thatthe optimum concentration for such growth is about1 X 10-2 At with the pollen used in this experiment,and that concentrations in excess of 4 X 10- 2

inhibit germination.\We next sought to determine what levels of cal-

cium were present in various tissues of snapdragon.how these levels were related to the chemotropicactivity of these tissues, and to what extent calciumwas released from tissues into agar. About 350flowers were collected and dissected into the partslisted in table III. All of these were analyzed for

Table IIIThe Concentration of Calcium in the Floral Tissucs

of Snapdragon

Tissue

Style, upper thirdStyle, middle thirdStyle, lower thirdOvules and placentaOvary w!\allsPedicelsSepalsPetalsFilaments

Dry wt

g0.1950.1350.1370.3320.5591.2983.06619.1522.520

Ca Activity*% dry xvt

0.510.501.242.171.311.241.510.190.36

+-+++++

* This column indicates whether or not a chemotropicresponse was obtained when the fresh tissues of com-parable flowers were tested in the bioassay apparatus.

calcium andl tested for chemotropic activity. Thedata show that there is a correlation between totalconcentration of calcium in the tissue and the chemo-tropic activity of the tissue. There is, moreover,

Table IIThe Efect of Calciuon (as CaCI?) in the Plate Medium on the Growth of Pollen Tutbcs Towards Ein ptv Test and

Control ChambcersThe plate medium contained 20%7 sucrose, 0.1% yeast extract, 0.017% boric acid, and 1%l agar. The readings were

made after incubation for 8 hours at 25°.

Growth to test chamber

Length (mm)

03. .

0.3

0.3-0.40.5-0.90.6-> 1.00.7-> 1.00.8-> 1.00.5-1.00.5-0.8No germinationNo germination

Growth to control chamber

Number

20

125120

-250-300

170148130

0

0

Length (mm)

0.3.

0.30.30.5-0.90.6-> 1.00.7-> 1.00.8-> 1.00.4-0.80.5-0.8No germinationNo germination

CalciumMolarity

01 x 10 C1 x 10-1 X 10-41 x 10 '5 x 10231 X 10-22 x 1023 x 10-24 x 10-25 x 10 21 x 101

Number

007

44150

-250-300150135105

00

72

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

MASCAREN HAS AND MACHLIS-CHEMOTROPIC RESPONSE

an overall gradient amounting to a fourfold increasein concentration between the upper third of the styleand the ovules and placenta. These data indicatethat there is no gradient down through the first two-thirds of the style. It must be remembered, however,that the entire stylar tissue was analyzed whereas thepollen tubes presumably grow only in the stigmatoidtissue (5). Although attempted, no way was foundto dissect the stigmiatoid tissue from the rest of thestyle.

An unsuccessful attempt was made to observe thedistribution of calcium by radioautography of longi-tudinal sections of the style. The gynoecia from aplant fed Ca45 were too heavily labeled to permitthe necessary observations. The radioactive gynoeciaand pollen were used, however, to establish the rela-tionship between the content of Ca45 and chemotropicactivity and the extent to which calcium diffused fromthe tissues into agar.

Ovules and placenta as a mass were removed from3 flowers whose anthers had just dehisced. Each ofthese masses was divided approximately in half. Oneof each half was assayed for chemotropic activity andthen dried, ashed, and counted. The results are givenin table IV. Although there is not a strict propor-

Table IVCornlarison of Ca" it Ovules* and Chemotropic Activity

Number of tubes Cpmto ovules*

73 4,762132 14,600120 6,250

* Including placental tissue.

tionality between radioactive calcium content andchemotropic activity, the 2 measures do vary in thesame direction. The other halves of the ovule andplacenta masses were used to determine the extentto which calcium is released. In place of the papershown in the upper diagram of figure 1, a piece ofdialyzing membrane was inserted, extending down-ward to the bottom of the agar and 1 mm beyondeach end of the chamber, to facilitate the subsequentremoval of the tissue. The tissue was placed in thechamber against this membrane and allowed to standfor 6 hours at 250. The ovules, placenta, and mem-brane were then removed, dried, ashed, and assayedfor radioactivity. The agar to the left of the mem-brane was removed as a block 3 mm wide, 7 mmlong (1 mm beyond each end of the chamber), andthe full depth of the agar. It too was dried, ashed, andassayed for radioactivity. The results are tabulatedin table V and indicate that almost 8% of the radio-active calcium moved from the tissue into the agar.This value is probably low because of the likelihoodthat not all of the agar containing Ca45 was cut out.In an earlier experiment 10% of the Ca45 was foundin the agar after a diffusion period of only 3 hours.

Pollen also contains calcium and readily gives itup to agar. Pollen collected from flowers grown in

Table VThe Release of Ca" from Ovules

Flower Counts per minute Ca released

No. Ovules* Agar S01 11,098 958 8.62 5,542 547 10.13 14,273 884 6.2Average 10,304 796 7.7

* Including placental tissue.

the Botanical Garden was found to contain 0.347%calcium on the dry weight basis. Pollen from a plantfed Ca45 was spread in a thin layer on a 1 square inchpiece of dialyzing membrane which was then laid onthe surface of a nutrient agar plate and left there for3.5 hours at 250. At the end of the diffusion periodthe membrane and pollen were removed and assayedfor activity as was the agar beneath the membrane.The pollen and membrane contained 4,531 counts perminute of calcium and the agar 432 counts perminute. Thus approximately 107% of the calciummoved from the pollen into the agar.

The next set of experiments was designed to sub-stantiate our conclusion that calcium is a chemotropicagent for the pollen tubes of snapdragon. The firstof these attempts was to distinguish between the en-hancement of growth per se and a chemotropic re-sponse. The bioassay method does not permit ob-servation of individual pollen tubes. Since it is knownthat calcium markedly enhances the growth of pollentubes (2) it could be argued that those tubes in thepollen chamber initially directed toward the testchamber grow faster than those initially directed tothe control chamber and hence are the only ones toregister within the time allowed for growth (normally8 hours). This cannot be clarified simply by ex-tending the incubation period because sooner or latercalcium from the test compartment will diffuse throughthe pollen chamber and into the control area. Threeother substances are known to enhance the growthof pollen tubes: sucrose (8, 9, 10, 12, 17, 18) ; yeastextract (3, 7); and boron ( 11 ).

It was shown that the rate of growth of the pollenused in this study was enhanced by sucrose, yeastextract, and boric acid. Pollen was sown on agarplates containing sucrose, yeast extract, boric acid,and CaC12 at optimal levels (107% sucrose, 0.17%yeast extract, 0.01% boric acid, and 1 X 10-2 MCaC12) excepting that nutrient whose effect on therate of growth was being tested. The concentrationof the latter (sucrose, yeast extract, or boric acid)was varied. The minimal concentration of sucrosetested was 27% since the complete absence of sucroseprevented germination; with yeast extract and boricacid the minimal concentration was zero because rea-sonable although reduced germination of pollen didoccur when these were not incorporated into the media.The ratio of the average length of the tubes in theoptimal concentration of each nutrient compared tothat in the minimal concentration was 3.8 for sucrose,

73

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

2.2 for yeast extract, and 5.2 for boric acid. When,now, sucrose or yeast extract gradients were estab-lished between the test chamber and the pollen cham-ber no pollen tubes in excess of the controls grewtowards the test chamber. In the experiments wraithsucrose, the plate medium contained 2% sucrose withthe other nutrients at optimal concentrations as speci-fied just above. \With yeast extract, the plate mediumwas devoid of yeast extract but with the other nu-trients at optimal levels. In both series of tests,calcium was supplied in the plate medium as CaC19at both suboptimal (2 X 10- M) anlc optimal(2 X 10-2 si). Thus, the enhancement of thegrowth of the pollen tubes by either yeast extract orsucrose is not involved in the chemotropic responseindicated by the assay method.

In the tests with sucrose and yeast extract gradientsit was assume(l that there xvas sufficient time (luringthe 8 hour incubation period for these substances todiffuse through the test wall and thus be capable ofaffecting the pollen tubes. Direct testing of thisassumption was not (one. However, the germina-tion of the pollen in the l)ollen chamber as well asthe average length of the pollen tubes growing intothe test wall were greater with the higher concen-trations of sucrose (10 and 15% ) than with the lowerones; with yeast extract, w\Nhich affects germinationless markedly than sucrose, the evidence for diffusionrests only on the greater lengths of the tubes in re-sponse to the higher concentrations of yeast extractin the test chamber.

In contrast to these negative results with sucroseand yeast extract are those obtained with concen-tration gradients of boric acid. The initial experi-ment is recorded in table VI. Boric acid in the testchamber unaccompanied by calcium is without effecteven though the plate medium contains calcium at aconcentration just below that which causes tubes togrow out of the pollen chamber in all directions inthe absence of anyi gradient of calcium. When thissame concentration of calcium (2 X 10- I) isadded together with boric acid to the test chamberthere is a significant increase in the number of tubesgrowing to the test chamber compared to the controlchamber. XWhen there is a gradient of calcium(1 X 10-2 Mr in the test chamber and 2 X 10 Orin the plate medium) the response to calcium isgreatly increased by the optimum concentration ofboric acid in the test chamber. A large number ofadditional experiments on this effect of boric acidwere performed. The concentration of calcium thatmust be present to detect the effect of boric acidvaries with the source of the pollen. With a dif-ferent collection of pollen than that used in the pre-ceding experiment no effect could be detected untilthe calcium concentration in the chamber and platemedia reached 1 X 10 \2r. With pollen collectedfrom commercially grown snapdragons no effect ofboron was found even with 1 X 10-2 AI calcium inthe media. The results of all the experiments indi-cate that boric acid enhances the chemotropic re-

Table VIThe Effect of Conicenitration Gradients of Boric Acid on

the Chemiotropic Response of Pollten TntbcsThe test media contained 10% sucrose, 1% agar, and

0.1% yeast extract in addition to the indicated amounts ofboric acid and CaCI2. The media were added to the testchamber as 30 ,ul drops which then solidified. The plateand control media contained 10% sucrose., 1%c agar, 0.17%yeast extract, and 2 X 10-'M CaCl2.

Medium in test chamber

Boric Acid

0.0000.0050.0100.0200.0300.0500.1000.0000.0050.0100.0200.0300.0500.1000.0000.0050.0100.0200.0300.0500.100

CaM

000000

2 x 10'42 x 10-'2 x 1042 x 10-42 x 10'12 X 10-4

10-210 210 2

10 210-2

10--

Number of tubes

Test Control

2115;2 12831

2824.0439465179185 IQ586191

1151305;6

2417192226363622162)2626211936182121274747

spouse to calcium, that in the absence of calcium ithas no chemotropic effect, and that the concentrationof calcium necessary for this effect of boric acid to bedetected depends on the physiological ( nutritional ?)status of the pollen.

The relationship between the sponitanieous growthof pollen tubes into the test and control walls of thebioassay was earlier examined (table II) with thefinding that a concentration of calcium in the platemedlium in excess of 2 X 10-4 \-was needed forsuch spontaneous growth (with the particular pollenuse(l) and that 1 X 10 2 M was optimum for thegrowth of the tubes out of the pollen chamber in alldirections in the absence of a gradient of calcium.In these tests the plate medium contained 0.01 %o boricacid. The finding that boric acid effects the responseto calcium necessitated a testing of the appropriatenessof the concentration of boric acid used in the earlierexperiments. Plates w\vere therefore prepared withthe usual nutrients, 2 X 10-2 At CaC12,, anc dif-ferent concentrations of boric acid. The maximumnumber of tubes grew toward the test and controlchambers, which were empty, wvhen the boron con-centration in the medium was 0.01 %c. This resultis another indication of an interaction between boricacid and calcium and confirms the use of 0.01% boricacid as optimal in all the preceding experiments.

To further confirm that calcium was exerting achemotropic effect, tests were made involving methodsother than the standard bioassav. In these tests

74

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

MASCAREN HAS AND MACHLIS-CHEMOTROPIC RESPONSE

CaCO3 and later, platelets of calcium DL-malate wereused since they made it possible to maintain a con-stant concentration of calcium at the source becauseof their low solubility. It was presumed that thismore closely simulated the conditions prevailing whentissue was used as the source of the chemotropic fac-tor. When CaCO3 was used for the standard bio-assay, it was introduced into the test depressionagainst the inner wall as the powder approximately4 to 5 hours prior to introducing the pollen. Withthe malate, no time difference was necessary foroptimum results and the malate was introduced as asheet laid against the inner wall of the test compart-ment.

Figure 2 shows a series of tests conducted withCaCO3 and the comparisons of the responses observedwith those elicited by ovules. Figures 2a and hillustrate the growth of pollen tubes in the standardassay apparatus to CaCO3 and to ovules placed in theleft depression. No additions were made to the con-trol depressions. In figures 2c and d the pollenwas spread in a row along the edge of each side ofthe pollen depression. Many tubes grew into thewall towards the test depressions which containedCaCO3 or ovules whereas very few grew into theopposite wall. In figures 2e and f pollen grainswere placed in a slit in the agar approximately 1 mmfrom the long outer edge of the test depressions inwhich CaCO3 and ovules had been placed. Finally,in figures 2g and h the traditional method of display-ing pollen tube chemotropism was used. CaCO3 andovules were placed 1 mm away from pollen grainsspread on the surface of the agar. When ovules orCaCO3 are omitted from the types of tests illustratedby 2a, b, c, and d, the tubes tend to remain, but notcompletely, in the pollen chamber. When the attract-ants are omitted from tests 2e, f, g, and h, then thepollen tubes grow randomly in all directions.

These illustrations show that the pollen tubes re-spond to Ca as they do to ovules; however, the re-sponse is almost always less intense to the calciumthan to the ovules. In those tests where the pollengrains are spread in rows, it is possible to observethe individual tubes and to note that whatever thedirection of emergence of the tubes, most of them turnand by a convoluted pathway eventually grow to-ward the source of the chemotropic agent whether itis ovules or calcium.

An attempt was made to see if ovules would attractpollen tubes in the presence of calcium. The pro-cedure used was the usual method in which ovulesare placed on the surface of a nutrient agar platewith the pollen placed individually about 1 mm fromthe ovules. Although we never were able to obtainfully consistent results with this method [see discus-sion in (16)], the usual spectacular response as shownin figure 2g was obtained in more than half of the tests.When the agar contained 2, 3, and 4 X 10-2 CaC12no such chemotropic effect was detected in 31 trials.Thus, adequate calcium in the plate medium erasesthe chemotropic effect of ovules.

The pollen of Narcissus pseudonarcissus andClivia, miniata also responded to calcium in the bio-assay apparatus in the same manner as that of snap-dragon.

DiscussionA variety of evidences have been presented lead-

ing to the conclusion that calcium exerts a chemotropiceffect on the pollen of snapdragon. A comprehensivereview of the literature on pollen tube chemotropism(16) cites all work done on pollen tube chemotropismand is the basis for the following discussion.

The chemotropic factor is widely distributed inplant tissues or else there are several factors to whichpollen tubes can react tropically. Chemotropism ofpollen tubes of one species to nongynoecial tissues ofthe same and other plant species has been repeatedlyobserved as has chemotropism between pollen andgynoecial tissue from different plant species, evenwhen the plants were widely separated taxonomically.Thus the chemotropic substance is of common occur-rence and not a very specific compound. Positivechemotropic responses to specific compounds andcrude preparations were obtained by the early work-ers-to compressed yeast, sucrose, glucose, fructose,lactose, diastase, and egg albumin. However,although more recent work has not confirmed theseresults, they cannot be altogether discounted. It ispossible that unlike the compounds obtainable at thepresent time, the compounds used in the 1890's con-tained impurities that were chemotropic, probably cal-cium. It is known that diastase which is one of thematerials earlier found to have chemotropic activitydoes contain -one or more gram atoms of calciumper mole of enzyme (6).

The chemotropic factor (s) is known to be heatstable, resistant to dilute acid hydrolysis, small enoughto pass through a dialysis membrane, and soluble inwater and alcohol, but not in acetone or ether. Cal-cium possesses all these properties.

The active substance from gynoecial tissues isknown to have in addition to a tropic effect also agrowth-stimulating effect on pollen tubes. Whenpollen is sown in sufficiently large colonies on nu-trient agar, growth begins in the center and thetubes grow towards the periphery and go a short dis-tance beyond the limits of the colony. Then withonly a few exceptions, the tubes recurve towards thecolony. Sometimes tubes that grow beyond the limitsof a colony grow into a neighboring colony. Evi-dently there is some substance within the colonieswhich influences the direction of growth of the pollentubes. As the number of grains in the colony in-crease, the pollen germination percentage and tubelength proportionately increase. Recent work byBrewbaker and Kwack (1) has shown that the factorcausing this population effect is calcium.

Calcium, which has been found to have a tropiceffect in addition to a growth effect, appears to pos-sess all the properties of the tropic substances forpollen tubes that have been reported in the literature.

The conclusion that calcium is a chemotropic fac-

75

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

PLANT PHYSIOLOGY

*1:

.,A

*i.7.3

a*:il4. ":?.. o '* S:..t .:.

^ S: :|:m ,!X .rw

S ^.,#

- .:...

F. # o:.'*s

*:: xti L,,tS w t #

t t '.

.. C4|E_....

...:R

* S ....s.

;..Xf a

11M w

.:

:s..

....:

...

..# w; . 8-. vi. .,|f,,3F .. ^

_ 't:.9iX iv.s _ JjX

t

v.: ::.f.t

wa ~~~~~~~~~...... ........

.....~~~~~~~~~~~~~~~~~~~....

FIG. 2. The chemotropic response of pollen tubes to ovules (left) and to CaCO3 (right). a and b: Standard bio-assay. c and d: The pollen grains were spread in a single-grain row along each edge of the pollen chamber. e and

76

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

MASCARENHAS AND MACTILl -CHEMOTROPIC RESPONSE

tor is further supported by the distribution of calciumin floral parts, the readiness with which calcium dif-fuses from pollen and floral parts into agar, and theinability to obtain a chemotropic response by ovules inthe presence of ample calcium. There was shownto be a very substantial overall gradient in calciumfrom the stigma to the ovules. As pointed out earlier,the lack of a gradient in the upper part of the stylemay not reflect the actual distribution of calciumaffecting the pollen tubes in their growth throughthe style because they are presumably restricted to thestigmatoid tissue. The actual situation in snap-

dragon was not investigated. It would be desirableto work with a plant where the distribution in stigma-toid tissue itself could be examined. Although we

were unable to observe chemotropisnm to ovules in thepresence of calcium using the surface test, Brewbakerand Kwack state that they do get a response in theirtest procedure (2).

The ready diffusion of calcium from pollen grainsoffers a possible explanation for both the recurvatureof pollen tubes back to a colony of pollen referred to

above and for the failure of pollen tubes to grow out ofthe pollen chamber in the bioassay apparatus unlessthe plate medium contains high levels of calcium.In both cases the released calcium could be presumedto act chemotropically. Quantities of pollen, all re-

leasing calcium into the agar, would result in a local-ized concentration of available calcium tending to

keep the pollen tubes in the vicinity.Evidence was presented that the optimum response

of pollen tubes to calcium was conditioned by the pres-

ence of boron, depending on the particular lot ofpollen used in the tests. Although calcium elicits a

response in the bioassay plate in the absence of addedboron, the reverse does not happen. Not enough isknown about this interaction to explain it.

Summary

Evidence is presented that the pollen tubes of snap-

dragon (Antirrhinum niajus) grow chemotropicallytowards a source of calcium. Analyses of the partsof the gynoecia of snapdragon indicate the highestconcentration of calcium to be in the ovules. Boricacid enhances the chemotropic response to calcium.

Acknowledgment

We wish to thank Miss Linda Simons for her technicalassistance.

Literature Cited1. BREWBAKER, J. L. AND S. K. MAJUMDER. 1961.

Cultural studies of the pollen population effect and

the self-incompatibility inhibition. Am. J. Botany48: 457-64.

2. BREWBAKER, J. L. AND B. H. KWACK. 1963. Theessential role of calcium ion in pollen germinationand pollen tube growth. Am. J. Botany 50: 859-65.

3. BRINK, R. A. 1924. The physiology of pollen. IV.Chemotropism; effects on the growth of groupinggrains; formation and function of callose plugs;summary and conclusions. Am. J. Botany11: 417-36.

4. CARLSON, R. M. AND C. M. JOHNSON. 1961. Chelo-metric titration of calcium and magnesium in planttissue. Method for the elimination of interferingions. J. Agr. Food Chem. 9: 460-63.

5. ESAU, K. 1953. Plant Anatomy. John Wiley andSons, Inc., New York. 735 pp.

6. FISCHER, E. H., W. N. SUMERWELL, J. JUNGE, AND

E. A. STEIN. 1958. Calcium and the molecularstructure of x-amylases. Proc. 4th Internatl. Cong.Biochem., Vienna. 8: 124-37.

7. GLENK, H. 1960. Keimversuche mit Oenothera-Pollen in vitro. Flora 148: 378-433.

8. HRABETOVA, E. AND J. Tupy. 1961. Respiration ofapple pollen on different sugar substrates and theproblem of the role of sucrose in pollen-tube growth.Biol. Plantarum (Praha) 3: 270-76.

9. IWANAMI, Y. 1956. Physiological researches of

pollen. IX. The starch and the sugar in the pollengrain (Japanese with English summary). Botan.

Mag. Tokyo 69: 91-5.10. IWANAMI, Y. 1956. Physiological researches of

pollen. X. The absorption of sugars and some

enzyme reactions (Japanese with English sum-

mary). Botan. Mag. Tokyo 69: 198-202.11. JOHRI, B. M. AND I. K. VASIL. 1961. Physiology

of pollen. Botan. Rev. 27: 325-81.12. LINSKENs, H. F. 1958. Zur Frage de Entstehung

der Abwehr- Korper bei der Inkompatibilitats-Reaktion von Petunia. I. Mitteilung: Versuchezur Markierung der Griffel mit P.,-und C"1-Ver-bindungen. Ber. Deut. Botan. Ges. 71: 3-10.

13. MACHLIs, L. AND J. G. TORREY. 1956. Plants inAction. W. H. Freeman and Co., San Francisco.282 pp.

14. MASCARENHAS, J. P. AND L. MACHLIS. 1962. Thepollen-tube chemotropic factor from AWtFrrhinumniajus: bioassay, extraction, and partial purification.Am. J. Botany 49: 482-89.

15. MASCARENHAS, J. P. AND L. 'MACHLIS. 1962.Chemotropic response of Antirrhinum majors pollento calcium. Nature 1962: 292-93.

16. MASCARENHAS, J. P. AND L. MACHLIS. 1962. The

hormonal control of the directional growth of pollentubes. Vitamins and Hormones 20: 347-72.

17. O'KELLEY, J. C. 1955. External carbohydrates in

growth and respiration of pollen tubes in vitro.Am. J. Botany 42: 322-26.

18. Tupy, J. 1960. Sugar absorption, callose formationand the growth rate of pollen tubes. Biol. Plan-tarum (Praha) 2: 169-80.

f: The pollen grains were inserted in a row in a slit made in the agar 1 mm from the outer edge of the control cham-ber. The test materials were placed against the outer wall of the control chamber. g and h: The pollen grainswere placed on the surface of nutrient agar and the test materials 1 mm awav.

77

I

Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.