Determination of the Cellular Mechanisms Regulating Thermo

Plant Physiol. (1991) 97, 630-6370032-0889/91 /97/0630/08/$01 .00/0

Received for publication January 15, 1991Accepted June 12, 1991

Determination of the Cellular Mechanisms RegulatingThermo-induced Stem Growth in Thiaspi arvense L.

James D. Metzger* and Kendall DusbabekUnited States Department of Agriculture, Agricultural Research Service, Biosciences Research Laboratory,

State University Station, Fargo, North Dakota 58105

ABSTRACT

Field pennycress (Thiaspi arvense L.) is a species with a coldrequirement for the initiation of reproductive development (ther-moinduction). Work in this laboratory has been focused on elu-cidating the biochemical and molecular mechanisms underlyingthe bolting or rapid stem elongation response that is an intricatepart of reproductive development in this species. In the presentpaper the cellular basis for thermo-induced stem growth wasdetermined. Evidence is presented indicating that bolting resultsfrom the production of new cells that elongate to their originallength before thermoinduction. This increase in cell division oc-curs in the pith and cortex approximately 0.5 to 5.0 millimetersbelow the stem apex. For at least the early stages of thermo-induced stem growth, enhanced cell elongation does not appearto be a factor because average lengths of pith cells from stemsof thermo-induced plants were similar or less than noninducedcontrols. In addition, both the amount of increase in the productionof new pith cells and stem growth were positively correlated withthe length of the cold treatment. Two other lines of evidence arepresented corroborating previous assertions (JD Metzger [1985]Plant Physiol 78: 8-13) that gibberellins mediate thermo-inducedstem growth in field pennycress. First, treatment of noninducedplants with gibberellin A3 completely mimicked the effects of a 4-week cold treatment on mitotic activity in the pith and cortex.Second, very little increase in the production of new cells wasobserved in the pith and cortex of thermo-induced plants of agibberellin-deficient dwarf mutant of field pennycress. It is alsoshown that the influence of photoperiod on stem growth is me-diated by an effect on the final length that cells ultimately attain.

It is well known that environmental factors play an impor-tant role in the regulation of of stem growth. Some of themost dramatic responses to environmental stimuli occur inrosette plants in which internode expansion is minimal untilthe plants are subjected to changes in specific environmentalconditions such as daylength or temperature. The inductivestimuli trigger a sequence of events that result in rapid stemgrowth (bolting). At least initially, the increase in stem growthis the result of expansion of internodes that existed before theinductive treatment. The bolting response is an integral partof reproductive development and is closely associated withfloral initiation (15, 26).

In many instances, GAs' are an obligate part of the signaltransduction pathway leading to bolting, and although the

' Abbreviation: GA, gibberellin.

molecular details of GA involvement are not yet clear, itseems likely that alterations in GA biosynthesis and metabo-lism induced by the stimulus have an important role (15, 26).In spinach, for example, rapid stem growth resulting from alengthening of the photoperiod is associated with dramaticchanges in endogenous GA levels and metabolism (3, 20).Moreover, certain enzymes in the GA biosynthetic pathwayof this species have been shown to be light regulated (4, 5).Our work in this area has centered on elucidating the

biochemical and molecular events that underlie thermo- orcold-induced stem elongation in field pennycress (Thlaspiarvense L.). This species has a requirement for a period oflow temperature (0-1 5°C) before stem elongation is initiated(14). Previous results demonstrated that thermo-induced stemgrowth in field pennycress is mediated by GAs (14). Morerecently, we (7) showed that metabolism of the GA precursor,ent-kaurenoic acid is under thermoinductive regulation in theshoot tips, the site of perception of low temperatures (17), butnot in the leaves. We further hypothesized that increasedconversion of this compound to GAs following thermoinduc-tion forms the biochemical basis for thermo-induced stemgrowth in field pennycress (7, 18).

However, the correlation between specific changes in GAmetabolism and macroscopically measurable increases instem growth is not a straightforward indication of a cause andeffect relationship because stem growth is the product of cellelongation and the production of new cells. Thus, the assign-ment of a causal role for a specific alteration in GA metabo-lism could be erroneous if it occurred between the onset ofmeasurable stem growth and the cellular events that form thebasis of this process. The purpose of the research described inthis paper was to determine the cellular basis for thermo-induced stem growth in field pennycress and to compare thetiming of cellular changes with other thermo-inducedprocesses.

MATERIALS AND METHODSPlant MaterialWith one exception, all experiments were performed with

the CR, inbred line of field pennycress (Thlaspi arvense L.)grown under identical photoperiodic and temperature condi-tions as described before (14, 16). In one experiment, plantsof the single-gene GA-deficient dwarf mutant derived fromCR, (EMS-141) were used (19). Unless otherwise noted, 6-week-old plants were thermo induced by subjecting them to6°C for 4 weeks (14). Noninduced plants used in these studies

630

CELLULAR BASIS OF STEM GROWTH IN FIELD PENNYCRESS

were grown at 21C and, at the time of use, were the samechronological age as the thermo-induced plants.

Before and during the cold treatment, plants were grownunder SD conditions consisting of 8 h of high intensity lightfrom fluorescent and incandescent lamps (14, 16). Unlessotherwise noted, plants were subjected to LD immediatelyfollowing the cold treatment. Noninduced controls were alsotransferred to LD at this time. Daylength was extended with16 h of low intensity light from incandescent lamps followingthe 8-h light regimen described for SD (14, 16).Stem elongation was promoted in 8- to 10-week-old non-

induced plants by applying 10 gL of an aqueous solutioncontaining 10% (v/v) acetone, 0.05% (v/v) Tween-202, and10 ,ug GA3 (Sigma) to the shoot tips on alternate days fora total of six applications. Control plants received similartreatments except that no GA3 was included in the appliedsolution (14).

MicroscopyAll plants used for this study were harvested at 16:00 h.

Stems, with all leaves >2 mm long removed, were dividedinto 1-cm segments and then fixed in an ethanol:water:formalin:acetic acid mixture (5:3.5:1:0.5, v/v, respectively).The stem segments were then dehydrated in a graded water-tert-butanol series and embedded in paraffin (9). Longitudinalserial sections, 8 gm thick, were prepared from the transverseaxis of the stem using a rotary microtome, mounted on glassslides, and then stained with safranin and fast green (9).The number of mitotic figures observed in the stained

sections was used as a relative indication for the presence ofdividing cells in various parts of the stem. The median eightsections of each segment were examined with a light micro-scope for the number of nuclei undergoing mitosis. All stagesofkaryokinesis were included in the mitotic counts. However,mitotic figures in early prophase were difficult to see, andtherefore, many nuclei at this stage of mitosis were probablymissed. The data are expressed as the sum of the mitoticfigures observed in all eight sections (64Mim total) as a functionofdistance from the stem apex. Data for segments longer than500 Mm are reported as the sum of component sections.The lengths of pith cells in the median longitudinal section

for each stem segment were determined using a Leitz ocularmicrometer. Fifty cells were randomly chosen for measure-ment in each 500-,um segment beginning at the apex.The data reported are the results from one replicate. How-

ever, all experiments had two replicates and were repeated atleast one time with similar outcomes.

Growth MeasurementsThe length and the number of pith cells of the first four

elongating internodes in bolting plants were monitored. Fromstained longitudinal sections of stems, the distance betweentwo adjacent nodes was determined. Because leaves of field

2 Mention oftrademark or proprietary product does not constitutea guarantee or warranty of the product by the U. S. Department ofAgriculture and does not imply its approval to the exclusion of otherproducts that may also be suitable.

pennycress are arranged in a 2/5 phyllotaxy, there are fourinternodes between two adjacent nodes in a longitudinalsection. Thus, dividing the distance between them by fourprovides an average length ofthe four internodes. The averagenumber of pith cells along the transverse axis of the first fourelongating internodes was calculated by first counting thenumber of cells in a 1-mm region located between two adja-cent nodes and then multiplying that number by the averageinternode length.

Macroscopically visible internode growth induced by ther-moinduction or GA3 treatment (bolting) was measured witha ruler as described before (14). It should be noted that thismeasurement does not represent the true stem length and,therefore, should not be compared directly to data concerningthe location of mitotic figures or mean cell lengths expressedas a function of distance from the apex.

RESULTSLocalization of Thermo-Induced Stem Growth

Field pennycress plants grown for 6 weeks at 21C (thetypical age of plants at the beginning of a 4-week cold treat-ment) achieved stem heights, as measured from the apex tothe oldest true leaf, of about 12 to 15 mm. During the sameperiod, approximately 50 to 55 leaves 1 mm or more in lengthwere produced. Internodes on a rosette plant are, therefore,extremely short, and it is in fact impossible to measure theirlengths with a ruler because they are totally obscured by thebases ofthe petioles. Nevertheless, an average internode lengthin rosette plants was calculated from the stem height and leafnumber to be in the range of 250 to 350 Mm.

Internode elongation (bolting), measurable with a ruler, isusually first observed in about 7 or 8 d following the end of a4-week thermoinductive cold treatment (14). Elongation be-gins in the internodes bracketed by leaves 25 to 30, countingfrom the oldest true leaf. This region is located between 7 and9mm below the apex. Other internodes soon begin to elongatein a sequential fashion to progressively younger ones.

Effect of Thermoinduction on Cell DivisionMicroscopic examination of transverse stem sections

showed dramatic morphological changes in the pith followingthermoinduction. The pith ofnoninduced plants is comprisedof rounded cells arranged in a random fashion. Within 4 dafter the end of a 4-week cold treatment, pith cells wererectangle shaped and arranged in long linear files between 1and 4 mm below the apex. In subsequent days, the region oflinearly arranged pith cells expanded rapidly to longer dis-tances down the stem. However, the lower boundary of thelinear arrays of pith cells was always associated with firstinternodes to expand during thermo-induced stem growth(data not shown). These observations indicated that ther-moinduction results in extensive cell division in the pith.Evidence supporting this was obtained by using the numberof mitotic figures observed in transverse sections of stems asan indicator of the extent and locale of mitotic activity. A 4-week cold treatment resulted in a dramatic increase in thenumber of mitotic figures observed in the pith of stems whenplants were returned to 21C (Table I). A similar effect oncortical cells was also seen (data not shown). This increased

631

METZGER AND DUSBABEK

mitotic activity coincided with the appearance of flower pri-mordia, 4 d after the end ofthe cold treatment (19) but severaldays before measurable internode elongation began (14). Ahigh level of mitotic activity was maintained for at least 1week before decreasing sharply 14 d after the end of the coldtreatment (Table I). Under conditions used in this study, thedecline in the number of mitotic figures coincided with theonset ofthe linear (most rapid) phase ofthermo-induced stemgrowth (14).

Mitotic figures were widely dispersed in the pith and cortexthroughout the length of the stem, although most were ob-served at distances below the apex of 500 ,um and greater.

Initially, most mitotic activity was localized in an area from1 to 4 mm below the apex. This distribution of activitygradually spread further down the stem to a maximum of 12mm at 10 d after the end of the cold treatment before sharplycontracting (Table I).The increase in the number of mitotic figures following

vernalization was observed only in sections made from thelongitudinal axis of the stem; cross-sections from eitherthermo- or noninduced plants contained very few mitoticfigures (data not shown).

Effect of Thermoinduction on Cell LengthTable II shows that thermoinduction did not influence the

maximum mean length of pith cells in growing internodes, atleast not for 14 d after the end of the cold treatment whenthe most rapid phase of stem growth was commencing (14).Moreover, as the time after the end of the cold treatmentincreased, the maximum mean cell length was attained furtheraway from the apex, i.e. there was an inverse relationshipbetween cell length and the amount of mitotic activity when

Table I. Effect of a 4-Week Thermoinductive Treatment (60C) onMitotic Activity in Sequential Regions of the Pith

Values represent the sum of mitotic figures observed in the centraleight sections (8 gm thick each) taken from wax-embedded stemsegments.

Distance below Days after End of Cold Treatment NoninducedApex 0 2 4 6 8 10 12 14 Control

mm0.0-0.5 0 0 0 2 0 0 0 0 00.5-1.0 00 11 6 9 0 5 0 21.0-1.5 00 37 20 9 14 7 0 21.5-2.0 00 38 23 31 14 14 1 42.0-2.5 00 52 19 47 24 17 8 02.5-3.0 00 53 25 23 14 16 4 03.0-4.0 0 0 70 37 39 41 27 30 04.0-5.0 2 44 48 16 16 14 05.0-6.0 0 23 27 20 6 86.0-7.0 0 4 13 10 10 87.0-8.0 1 22 1 3 4118.0-9.0 0 0 1 2 0 09.0-10.0 0 12 0 0

10.0-11.0 0 13 0 011.0-12.0 11 0 0

Total 0 0 263 204 260 214 122 74 8Internode growth (mm) 0 0 0 0 1 4 11 21 0

Table II. Effect of a 4-Week Thermoinductive Treatment (60C) onPith Cell Length in Sequential Regions of the Stem

Values represent the mean of 50 cells per 500-Am segments takenfrom the central wax-embedded section. For the sake of clarity, SDvalues have been omitted, but maximum deviation from any meanwas 20%. The sections were obtained from the same plants usedfor mitotic figure determinations in Table I.

Mean Cell LengthDistance below Days after End of Cold Treatment Noninduced

Apex Noninduced___

0 2 4 6 8 10 12 14 control

mm Af0.0-0.5 13 12 13 13 12 13 14 12 130.5-1.0 15 17 14 16 17 15 16 18 151.0-1.5 19 21 15 17 19 22 19 23 171.5-2.0 24 23 1 6 1 7 21 22 23 23 222.0-2.5 43 23 1 7 23 23 22 23 25 352.5-3.0 63 32 1 9 24 24 26 25 26 573.0-4.0 65 53 20 23 24 28 25 27 684.0-5.0 62 37 25 26 27 29 27 765.0-6.0 47 26 26 27 31 306.0-7.0 61 28 27 28 32 367.0-8.0 33 32 28 40 408.0-9.0 37 33 26 57 589.0-10.0 57 60 32 61 59

10.0-11.0 64 67 35 64 6211.0-12.0 63 38 61 6325-30 60 68 62 6560-65 62 68160-165 61

comparing the two at distances greater then 500 gm belowthe apex (Tables I and II). For example, comparison of thenumber of mitotic figures and mean cell lengths for thesegments 2.5 to 3.0 mm below the apex from the varioussamples shows that there were 0, 0, 0, 53, 25, 23, 14, 16, and4 mitotic figures detected versus mean cell lengths of 57, 63,32, 19, 24, 24, 26, 25, and 26 ,um for noninduced, 0, 2, 4, 6,8, 10, 12, and 14 d after the end of the cold treatment,respectively (Tables I and II). Likewise, in the segments taken6 to 12 mm below the apex from plants harvested 10 d afterthe end of the cold treatment, only a modest number ofmitotic figures were observed, and the mean cell lengths wereintermediate (Tables I and II). This relationship betweenmitotic activity and cell length is undoubtedly due to the factthat mean cell lengths in a population will decline as theproportion of recently divided cells increases.

Relative Dependence of Internode Expansion on CellDivision and ElongationTable III shows the relationship between thermo-induced

internode expansion and cell division and subsequent elon-gation. The internodes chosen for analysis were the first fourto elongate during bolting. They were identified in transversestem sections by determining the position relative to the apexwhere linear files of pith cells that form following thermoin-duction ended. The number of pith cells in internodes ofnoninduced plants was always in the range of eight to 10,regardless of age or position of the internode. By 4 d after the

Plant Physiol. Vol. 97, 1991632


Table IlIl. Changes in the Average Internode Length and Number ofPith Cells in the Transverse Axis of the First Four Internodes toElongate following a 4-Week Thermoinductive Treatment at 60C

Days after End of Cold Treatment

0 4 6 8 10 12 14

Distance below 4 4 9 12 28 64 164apex (mm)

Mean internode 0.50 0.45 0.75 1.50 1.55 1.54 1.52length (mm)

Mean cell No. 8 22 24 29 28 28 27

end of the cold treatment, the average number of pith cells inthe first four expanding internodes increased from eight to 22cells per internode (Table III). This region, which was locatedbetween 3 and 4 mm below the apex, also contained a largenumber of mitotic figures (Table I). Moreover, the decreasein the average length of pith cells in this region of the stemfrom 65 ,um at 0 d to 20 ,um at 4 d after the end of the coldtreatment (Table II) was almost matched by the increase inthe average number of cells per internode.

In subsequent days, the mean internode cell number in-creased slightly to 27 to 29 cells per internode (Table III), andthis was correlated with few or no mitotic figures observed inthe internodes (Table I). In contrast, internode length in-creased more than threefold between 4 and 8 d after the endof the cold treatment. During the same period, the average

length of pith cells in the expanding internodes increasedproportionately with internode growth; these cells elongatedmore or less to the same lengths as those located 2.5 to 5.0mm below the apex of noninduced plants (Table II and III).

Previous work demonstrated a positive relationship be-tween the duration of the thermoinductive cold treatmentand several aspects of bolting (14). Increasing durations ofcold treatment resulted in greater stem growth rates, finalheights, and shorter times before the onset of measurablethermo-induced stem growth. Therefore, the effect ofdifferentdurations of cold on length and the number of pith cells inexpanding internodes was examined. Field pennycress plantswere subjected to 0, 2, 4, or 6 weeks of cold (6°C). At varioustimes after the end of the cold treatments, longitudinal sec-

tions of stems were prepared and the average number of pithcells in the first four elongating internodes of bolting plantswas determined as described above. Increasing duration ofcold treatment resulted in longer internodes and more pithcells per internode (Table IV). Furthermore, the increase ininternode length was entirely due to increased cell numberbecause average cell lengths were the same in all treatments.Interestingly, the proportional changes in cell numbers re-

sulting from the different cold treatments closely paralleledthose observed previously for thermo-induced stem growth(Table IV).

Increasing duration of cold also decreased the amount oftime necessary for the internodes to attain the maximumnumber of pith cells (Table IV), and these times were closelycorrelated with the time of initiation of thermo-induced stemgrowth following the various cold treatments (14). This is notsurprising because the attainment of maximum cell numberwas associated with internode expansion (Table III). Never-

theless, the first internodes to elongate were located 7 to 9mm below the apex for all three thermoinductive treatments.

Role of GA in Pith Cell Division and Elongation

The cellular effects of thermoinduction were, in general,mimicked in noninduced plants by exogenous GA3. Two daysafter the first treatment, linear arrays of rectangle shaped pithcells appeared 1 to 3 mm below the apex. The region of thepith containing the linear files of cells rapidly expanded duringthe subsequent treatment period, and as was the case forthermo-induced plants, its lower boundaries were closely as-

sociated with the first internodes to expand (data not shown).A modest increase in the number of mitotic figures was

observed 2 d after the first application of GA3. The numberof mitotic figures peaked dramatically at 6 d, followed by a

sharp decline thereafter despite repeated treatments (TableV). This peak in mitotic activity preceded the onset of meas-urable GA-induced stem elongation by 2 to 3 d. The distri-bution of mitotic figures in the pith of plants treated withGA3 was also similar to that observed in thermo-inducedplants, the bulk of which occurred 1 mm below the stem apex

(Tables I and V). Previous work showed that noninducedplants treated with 10 jig GA3 three times a week evoked a

growth pattern nearly identical with that induced by a 4-weekcold treatment (14). Consistent with this is the observationthat exogenous GA3 caused an increase in the number of pithcells in the first four elongating internodes from 8 to 27 to 306 d after the first treatment (data not shown). This is very

similar to the timing and the degree of increase in the numberof pith cells induced by a 4-week cold treatment (Table III).As was the case for thermoinduction, exogenous GA3 had

no effect on the maximum length that pith cells ultimatelyattained in the first 10 mm of the stem, and a similar negativerelationship between the number of mitotic figures and celllength was also observed (Table V). Therefore, as with ther-moinduction, GA-induced growth is preceded by a prolifera-tion of pith cells within the growing internode, followed byelongation to pretreatment values.Although exogenous GA3 elicited a similar response in

Table IV. Effect of Different Durations of ThermoinductiveTreatment (60C) on Various Parameters of Internode Growth

Measurements were made from longitudinal sections of the firstfour elongating internodes as described in "Materials and Methods."For purposes of comparison, the total amount of stem growth inducedby the various treatments observed in previous work (11) is shown.

Duration of Cold Treatment (weeks)

0 2 4 6

Mean internode length (mm) 0.54 0.85 1.82 2.23Mean cell No. 8 13 28 36Days to maximum cell No. 12 8 6Mean cell length (1km) 68 65 65 62Total thermoinduced stem 0 242 524 617

growth (mm)aa Total thermoinduced stem growth represents the distance from

the leaf subtending the first visibly elongating internode to the lasttrue leaf in mature (nongrowing) plants.

633


Table V. Effect of Exogenous GA3 on Mitotic Activity and Cell Length in Sequential Regions of the Pithof Noninduced Plants

Values represent the sum of mitotic figures in eight sections, and the numbers in parentheses equalthe means of cell lengths of 50 cells per 500 Mm in the central section of wax-embedded stem segments.For the sake of clarity, SD values have been omitted, but maximum deviation from any mean was 20%.

No. of Mitotic Figures (Mean Cell Length in Mm)Distance below Days after first application

Apex _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

0 2 4 6 8 10 12

mm0.0-0.5 0 (19) 0 (17) 1 (21) 0 (20) 0 (22) 10 (20) 2 (19)0.5-1.0 0 (21) 14 (19) 3 (19) 0 (25) 1 (23) 8 (20) 7 (20)1.0-1.5 0 (26) 10 (22) 23 (18) 4 (24) 11 (20) 7 (25) 6 (19)1.5-2.0 0(36) 5 (24) 20 (26) 6 (25) 11 (23) 7 (25) 9 (21)2.0-3.0 0 (59) 1 (36) 12 (32) 42 (24) 24 (20) 32 (23) 1 (24)3.0-4.0 0 (67) 1 (54) 2 (34) 22 (24) 39 (27) 23 (25) 1 (27)4.0-5.0 0 (65) 0 (58) 29 (25) 14 (25) 9 (27) 2 (33)5.0-6.0 0 (66) 62 (25) 13 (28) 6 (30) 0 (38)

6.0-7.0 27 (32) 10 (34) 0 (33) 0 (54)7.0-8.0 38 (36) 1 (40) 0 (51) 0 (64)8.0-9.0 10 (41) 0 (57) 0 (69) 0 (68)9.0-10.0 0 (70) 0 (64) 0 (67) 0 (66)

Total mitotic figures 0 31 61 240 124 102 28Internode growth (mm) 0 0 0 1 6 18 32

mitotic activity as thermoinductive temperatures, the twotreatments differed markedly in their effects on the timing offlower initiation. Following a 4-week cold treatment, micro-scopic flower primordia appear about 4 d after the end of thecold treatment (19), coinciding with the beginning ofthe peakof mitotic activity (Table I). However, flower primordia didnot become evident in the GA3-treated plants until 10 d afterthe first treatment, which was well after both the peak ofmitotic activity was observed and the initiation of bolting.The effect of reduced endogenous GA content on thermo-

induced cellular changes in the pith was examined using a

single-gene recessive dwarf mutant of field pennycress (EMS-141) that is GA deficient (19). In this experiment 6-week-oldplants ofEMS- 141 were thermo induced for 4 weeks and thenexamined microscopically at various times after the end ofthe cold treatment. As shown in Table VI, the severe effectsof reduced endogenous GA levels on thermo-induced stemgrowth in the dwarfwere associated with a profound reductionin mitotic activity in the pith. Likewise, fewer mitotic figureswere observed in the cortex as well (data not shown). Incontrast, pith cell lengths were similar or greater (especiallylower down the stem) than in the wild-type plants (Tables III

Table VI. Effect of a 4-Week Thermoinductive Treatment (60C) on Mitotic Activity and Cell Length inSequential Regions of the Pith in a GA-Deficient Dwarf Mutant (EMS-141) of Field Pennycress

Values represent the sum of mitotic figures in eight sections (8 !sm thick each), and the numbers inparentheses equal the means of cell lengths of 50 cells per 500 Am in the central section of wax-embedded stem segments. For the sake of clarity, SD values have been omitted, but maximum deviationfrom any mean was 22%. No internode expansion was observed in plants from any of the harvestdates.

No. of Mitotic Figures (Mean Cell Length in fim)Distance below Days after end of cold treatment

Apex2 4 6 8 10 12

mm0.0-0.5 0 (19) 0 (18) 0 (17) 0 (19) 0 (19) 0 (18)0.5-1.0 0 (19) 0 (20) 0 (20) 0 (19) 0 (21) 0 (20)1.0-1.5 0 (28) 0 (32) 4 (31) 0 (36) 0 (31) 0 (29)1.5-2.0 0 (39) 2 (45) 4 (48) 0 (45) 3 (45) 0 (40)2.0-3.0 0 (51) 6 (65) 2 (62) 2 (67) 0 (77) 4 (59)3.0-4.0 0 (78) 0 (84) 2 (86) 0 (88) 0 (84) 2 (76)4.0-5.0 0 (174) 0 (125) 0 (141) 0 (156) 0 (151) 0 (101)

Total mitotic figures 0 8 12 2 3 6

634 Plant Physiol. Vol. 97, 1991


and VI). Similar results were obtained when the GA biosyn-thesis inhibitor, 2-chlorocholine chloride, was applied tothermo-induced CR, field pennycress plants (data not shown).

Associated with a reduction of mitotic figures, the dwarfexhibited far fewer pith cells per internode than the wild type.Internodes from noninduced plants taken 4 mm below theapex contained four or five pith cells, and this number in-creased to only 8 to 10 at 14 d after the end of the coldtreatment (data not shown).

Effect of Photoperiod on Cell Division and Elongation

Previous work showed that photoperiod has a strong influ-ence on thermo-induced stem growth in field pennycress (14).Both thermo- and GA-induced stem growth are synergized byLD (14). It was of interest to examine the cellular basis forthe differences in thermo-induced stem growth under LD or

SD. Table VII shows that both the timing and the amount ofmitotic activity in pith cells were similar under either photo-periodic regimen, although there were slightly more totalmitotic figures observed under LD. Likewise, no differencesin mean cell lengths were observed in the first 10 mm belowthe apex (Table VII).

However, differences in thermo-induced stem growth inplants subjected to either LD or SD are more apparent whenplants reach the linear phase of growth, 12 to 14 d after theend of the cold treatment (14), and therefore, it should beeasier to detect differences in cell lengths in mature internodes.To examine this, plants were thermo induced for 4 weeks andthen allowed to mature under either LD or SD. After 4 to 5weeks when growth ceased (14), internodes were taken frompositions at 0 (apex), 25, 50, 75, and 100% (base) ofthe lengthof the stem from the apex, sectioned, and examined micro-

scopically as described before. Because the same total numberof leaves are produced in thermo-induced plants under LDor SD (unpublished observations), internodes taken from thesame relative section of the stem were developmentally com-

parable. Table VIII shows that the mean cell lengths ininternode sections from the LD grown plants were consider-ably longer than those in corresponding sections from SD.There was no additional increase in the number of pith cellsin the basal elongated internodes over that observed 2 weeksafter the end of the cold treatment. Thus, the pith cells retainthe capacity to elongate extensively well after they lose theability to divide.

DISCUSSION

One of the primary objectives of the research described inthis paper was to identify the cellular processes that underliethermo-induced stem growth in field pennycress. As in otherrosette plants, extremely short internodes are continuouslyproduced until the appropriate environmental signal inducesa caulescent growth habit. An important feature of bolting,whether induced by vernalization or photoperiod, is thatinternodes that were formed before the inductive signal elon-gate first and, therefore, are responsible for the initial phasesof the increase in stem growth (22). In the case of fieldpennycress, these internodes were located about 7 to 9 mmbelow the apex.

The length that an internode ultimately attains is a productof the number of cells within the internode and their length.The most obvious morphological change in the expandinginternodes following thermoinduction was an increase in thenumber of pith cells (Table III). In addition, longer coldtreatments resulted in internodes with increased numbers of

Table VIl. Effect of SD Conditions after a 4-Week Thermoinductive Cold Treatment (60C) on MitoticActivity and Cell Length in Sequential Regions of the Pith

Values represent the sum of mitotic figures in eight sections (8 Am thick), and the numbers inparentheses equal the means of cell lengths of 50 cells per 500 m in wax-embedded stem segments.For the sake of clarity, SD values have been omitted, but maximum deviation from any mean was 19%.The data in Tables and IlIl were obtained simultaneously and, therefore, a direct comparison betweenLD treatments is possible.

No. of Mitotic Figures (Mean Cell Length in inm)Distance below Days after End of Cold Treatment

Apex2 4 6 8 10 12 14

mm0.0-0.5 0 (12) 4 (14) 6 (11) 0 (13) 0 (13) 0 (12) 0 (13)0.5-1.0 3 (14) 15 (18) 43 (12) 10 (13) 0 (16) 0 (13) 0 (15)1.0-1.5 4 (22) 30 (23) 60 (15) 12 (14) 5 (19) 5 (22) 0 (16)1.5-2.0 9 (38) 14 (23) 62 (16) 24 (16) 18 (22) 24 (23) 2 (18)2.0-2.5 5 (59) 9 (36) 58 (18) 46 (18) 14 (23) 40 (22) 15 (20)2.5-3.0 0 (64) 0 (60) 50 (28) 56 (23) 29 (23) 18 (23) 31 (20)3.0-4.0 25 (36) 16 (41) 28 (25) 6 (47) 5 (51)4.0-5.0 20 (40) 5 (54) 28 (36) 0 (53) 1 (57)5.0-6.0 0 (54) 0 (63) 4 (45) 0 (62) 0 (62)6.0-7.0 0 (63) 0 (62) 0 (61) 0 (67) 1 (69)7.0-8.0 0 (67) 0 (63) 0 (65) 2 (69)

Total mitotic figures 21 72 324 169 126 83 57Internode growth (mm) 0 0 0 0 1 3 7

635


Table Vil. Effect of Photoperiod on Pith Cell Length of Fully GrownStems

Transverse segments, 5 mm in length, were taken from fivelocations along the length of the stem beginning with the apex andincluding segments at 25, 50, 75, and 100 (base) % of the length ofthe stem from the apex. The segments were fixed, embedded in wax,and sectioned. The lengths of 50 randomly chosen pith cells per 500,gm of the central longitudinal section were measured for a total of500 cells for each segment. The numbers in parentheses representthe shortest and longest cell observed in each segment.

Mean Cell Length (jAm)Relative Distance from Photoperiodic treatment

ApexLD SD

0 (apex) 128 (93-243) 86 (57-126)25 160 (105-243) 134 (99-179)50 200 (166-238) 139 (85-189)75 196 (163-258) 71(56-85)100 (base) 190 (145-244) 65 (47-86)

Final stem height (mm) 602 205

pith cells that were proportionally similar to the effects on

thermo-induced stem growth (Table IV). However, the aver-

age lengths of pith cells in the internodes from thermo-induced plants were not different from those in noninducedplants (Tables II and IV).

This leads to the conclusion that, although increased celldivision following thermoinduction cannot contribute di-rectly to growth (10, 12), it does provide more cells thatelongate more or less to the same extent as those in nonin-duced plants during the early stages of bolting. Thus, theprimary cellular process that controls thermo-induced stemgrowth in field pennycress is cell division, although it can onlypotentiate the maximum length the stem will ultimately at-tain. Consistent with this was the observation that the dura-tion of the cold treatment affected the number of pith cellsin internodes proportionally to the effects on stem growth(Table IV).The increase in mitotic activity following thermoinduction

was restricted to a region beginning about 500 Mm below theapex and gradually spread down the transverse axis of thestem as growth progressed (Table I). Changes in mitoticactivity occurring in the apical meristem were not observed,although any mitotic figure with a spindle apparatus lying ina plane not closely parallel to the longitudinal axis ofthe stemwould go undetected. In any event, the results are consistentwith the view of dicot stem histogenesis first proposed byHarting (6) and later elaborated by Sachs and coworkers (22-25), which holds that the cells responsible for stem growthoriginate from mitotic activity in a meristematic region dis-tinct from the eu- or true apical meristem termed the subap-ical meristem. This meristem encompasses a broad region ofpith and cortical cells beginning about 100 ,um below the apexand extends considerably throughout the growing portion ofthe stem (22).

In other dicots the subapical meristem, not the eumeristem,appears to be the target tissue for GA-induced stem growth

(23-25). For two reasons, this appears to be true for fieldpennycress as well. First, application of GA3 to noninducedplants caused an increase in mitotic activity that was nearlyidentical with respect to degree and distribution as a 4-weekthermoinductive cold treatment (Table V). Second, no com-parable increase in the number ofmitotic figures was observedfollowing thermoinduction of a GA-deficient dwarf mutant(Table VI). The fact that GA and thermoinduction haveidentical target tissues and elicit the same cellular responseprovides support for our previous contention that GA me-diates thermo-induced stem growth in this species (14). Inter-estingly, GA-deficient dwarf mutants of other species alsoexhibit greatly reduced numbers ofcells in the pith and cortex,and this forms the morphological basis for the dwarf growthhabit (1, 3, 21). Thus, during the vegetative phase of the lifecycle when field pennycress plants are rosettes, they behaveas GA-deficient dwarfs.How thermoinduction or GA affect the regulatory controls

of cell division is unknown, but it must involve an accelera-tion of the cell cycle because increasing duration of coldtreatment resulted in the production of progressively morepith cells in a shorter period of time (Table IV). Indeed,exogenous GA3 reportedly reduces the length of the G' and Sphases of the cell cycle in the subapical meristems of dwarfwatermelon seedlings (1 1).

Plants grown under noninductive conditions continuouslyproduce new leaves. This means that, depending on thetreatment used to induce stem elongation (e.g. 2, 4, or 6 weeksof cold or exogenous GA3), the plants had different numbersof internodes when the treatments were begun. However, inall cases the responsive regions ofthe pith were located furtherthan 1 mm below the apex. The maximum distance from theapex that mitotic activity extended was also similar in alltreatments (Tables I and V). Moreover, the first internodes toexpand were always located between 7 and 9 mm below theapex. This indicates that GA sensitivity is developmentallyregulated and apparently related to the maturation of cells inprogressively older internodes. The biochemical basis for theloss ofGA sensitivity in older internodes is unknown but maybe related to a loss of receptor function and/or other aspectsof the GA signal transduction pathway. Whatever the reason,the developmentally related loss of the ability to divide inresponse to GA represents a mechanism that influences thefinal length of an internode. The number of cells within aninternode will be determined by the rate of cell division andthe length of time that this process proceeds. By restrictingthe time period in which cell division within an internodeoccurs, the total number of new cells produced in response tovernalization will be limited by the rate of cell division. Inthis way the increased rate of pith cell production resultingfrom progressively longer cold treatments (Table IV) canresult in longer internodes.

In certain tissues of other species, GA appears to regulategrowth partly or mostly by controlling the length that individ-ual cells ultimately attain (2, 10, 13). However, there was noevidence for a direct effect of GA on cell elongation (TablesV and VI). In fact, pith cells in the GA-deficient dwarfmutantelongated to nearly the same extent as those in mature inter-nodes of the noninduced wild-type plants under LD and wereactually longer than those from wild-type plants subjected to

Plant Physiol. Vol. 97, 1991636


SD (Tables VI and VIII). Similar observations were made inother species in which GA regulates growth by modulatingmitotic activity in the subapical meristem (23-25). In contrastto the effect of GA on cell division in these cases, GAinfluences growth via an effect on cell elongation in othertissues such as light-grown lettuce hypocotyls in which cellnumber is essentially constant when the bulk ofgrowth occurs

(10, 12). This indicates the existence of two different mecha-nisms of GA action in growth regulation, the operation ofwhich depends on the organ: one primarily for indeterminategrowth such as that exhibited by stems in which GA regulatesproduction of new cells and a second for growth of organs

such as hypocotyls in which growth in the light occurs afterthe formation of most or all of the cells. Thus, there may bemore than one GA signal transduction system within a plant,each being tissue specific.Whereas thermoinduction and GA exert their effects on

stem growth in field pennycress through modulation of cellnumber, photoperiod influences growth by altering the max-

imum length cells attain (Tables III and VIII). This indicatesthat LD and GA increase stem growth through differentmechanisms and demonstrates that cell division and elonga-tion have separate regulatory inputs. Consistent with this areprevious observations that LD cannot substitute for cold inthe induction of stem growth but does synergize GA-inducedgrowth in noninduced plants (14). An identical dichotomyalso exists in the cellular bases for GA- and LD-inducedpetiole growth in field pennycress as well (16). Moreover, thedaylength extensions must contain far red light to increaseboth petiole (16) and thermo-induced stem (unpublished re-

sults) growth, suggesting a common response to changes inlight quality that may occur, for example, when plants becomeshaded by competitors (8).The timing of the cellular events caused by thermoinduc-

tion have a strong bearing on the assignment of a cause andeffect role for specific changes in endogenous GA levels. Asharp increase in mitotic activity was observed 4 d after theend of a 4-week cold treatment. Therefore, if an increase inthe level of an endogenous GA is responsible for thermo-induced stem growth, it must occur before this time and bemaintained during the entire period of high mitotic activityfrom 4 to 10 d after the end of the cold treatment (Table I).Likewise, such changes must occur in the target cells of thesubapical meristem.

Similar considerations in timing also apply in metabolismstudies as well. The results of this study are presently beingused in this laboratory in the design of studies on thermo-induced changes in the levels of endogenous GAs and GAprecursors, as well as the regulation of GA biosynthesis inrelation to thermo-induced stem growth in field pennycress.

LITERATURE CITED

1. Abbe EL, Phinney BO (1942) The action of the gene dwarf inthe ontongeny of the stem in maize. Genetics 27: 129

2. Cosgrove DJ, Sovonick-Dunford SA (1989) Mechanism of gib-

berellin-dependent stem elongation in peas. Plant Physiol 89:184-191

3. Denna DW, Munger HM (1963) Morphology of the bush andvine habits and the allelism of the bush gene in Cucurbitamaxima and C. pepo squash. Proc Am Soc Hort Sci 82: 370-377

4. Gianfagna T, Zeevaart JAD, Lusk WJ (1983) The effect ofphotoperiod on the metabolism of deuterium-labeled GA53 inspinach. Plant Physiol 72: 86-89

5. Gilmour SJ, Zeevaart JAD, Schwenen L, Graebe JE (1986)Gibberellin metabolism in cell-free extracts from spinachleaves in relation to photoperiod. Plant Physiol 82: 190-195

6. Harting MG (1845) Recherches micrometriques sur le developp-ment des parties elementaires de la tige annuelle des plantesdicotyledoneers. Ann Sci Nat Bot 3e Sci 4: 210-279

7. Hazebroek JP, Metzger JD (1990) Thermoinductive regulationof gibberellin metabolism in Thlaspi arvense L. I. Metabolismof [2H]-end-kaurenoic acid and [14C]gibberellin Al2-aldehyde.Plant Physiol 94: 154-165

8. Holms MG, Smith H (1977) The function of phytochrome inthe natural environment. II. The influence of vegetation can-opies on the spectral energy distribution of natural daylight.Photochem Photobiol 25: 539-545

9. Johanson DA (1940) Plant Microtechnique. McGraw-Hill, NewYork

10. Jones RL (1983) The role of gibberellins in plant cell elongation.CRC Crit Rev Plant Sci 1: 23-47

11. Liu PB, Loy JB (1977) Action of gibberellic acid on cell prolif-eration in the subapical meristem ofwatermelon seedlings. AmJ Bot 63: 700-704

12. Metraux J-P (1987) Gibberellins and plant cell elongation. In PJDavies, ed, Plant Hormones and Their Role in Plant Growthand Development. Martinus-Nijhoff, Boston, pp 296-317

13. Metraux JP, Kende H (1984) The cellular basis of the elongationresponse in submerged deep-water rice. Planta 160: 73-77

14. Metzger JD (1985) Role of gibberellins in the environmentalcontrol of stem growth in Thlaspi arvense L. Plant Physiol 78:8-13

15. Metzger JD (1987) Hormones and reproductive development.In PJ Davies, ed, Plant Hormones and Their Role in PlantGrowth and Development. Martinus-Nijhoff, Boston, pp 431-462

16. Metzger JD (1988) Gibberellins and light regulated petiolegrowth in Thiaspi arvense L. Plant Physiol 86: 237-240

17. Metzger JD (1988) Localization of the site of perception ofthermoinductive temperatures in Thiaspi arvense L. PlantPhysiol 88: 424-428

18. Metzger JD (1990) Comparison of biological activities of gibber-ellins and gibberellin-precursors native to Thlaspi arvense L.Plant Physiol 94: 151-156

19. Metzger JD, Hassebrock AT (1990) Selection and characteriza-tion of a gibberellin-deficient mutant of Thiaspi arvense L.Plant Physiol 94: 1655-1662

20. Metzger JD, Zeevaart JAD (1980) Effect of photoperiod on thelevels of endogenous gibberellins in spinach as measured bycombined gas chromatography-selected ion current monitor-ing. Plant Physiol 66: 844-846

21. Reid JB, Murfet IC, Potts WC (1983) Internode length in Pisum.II. Additional information on the relationship and action ofLe, La, Cry, Na, and Lm. J Exp Bot 34: 349-364

22. Sachs RM (1965) Stem elongation. Annu Rev Plant Physiol 16:73-95

23. Sachs RM, Bretz CF, Lang A (1959) Shoot histogenesis: theearly effects of gibberellin upon stem elongation in two rosetteplants. Am J Bot 46: 376-384

24. Sachs RM, Bretz C, Lang A (1959) Cell division and gibberellicacid. Exp Cell Res 18: 230-244

25. Sachs RM, Lang A, Bretz CF, Roach J (1960) Shoot histogenesis:subapical meristematic activity in a caulescent plant and theaction of gibberellic acid and amo-1618. Am J Bot 47: 260-266

26. Zeevaart JAD (1983) Gibberellins and flowering. In A Crozier,ed, The Biochemistry and Physiology of Gibberellins, Vol 2.Praeger, New York, pp 333-374

637

Documents

Determination of the Cellular Mechanisms Regulating Thermo