BioscienceHorizons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Volume 10 2017 10.1093/biohorizons/hzx001

Research article

Changing Plastid Dynamics within Early Rootand Shoot Apical Meristem-Derived Tissueof A. thaliana†

Lawrence Bramham* and Kevin Pyke

School of Life Sciences, University of Warwick, Wellesbourne Campus, Warwick, CV35 9EF

*Corresponding author: School of Life Sciences, University of Warwick, Wellesbourne Campus, Warwick, CV35 9EF.Email: lawrencebram@hotmail.co.uk

Supervisor: Dr Kevin Pyke, School of Biosciences, University of Nottingham, Sutton Bonington LE12 5RD, UK.

Whilst plastids are fundamental to many aspects of plant biology and the production of enhanced crop cultivars, researchinto the dynamics of non-green plastids has remained somewhat disregarded by the scientific community compared tochloroplasts. They are equally pivotal to normal plant development however, and are now increasingly becoming the focusof research made possible by genetic manipulation and reporter gene constructs.

The total plastid content of all plant cells originates from small, undifferentiated plastids termed proplastids found withinthe meristematic regions of both root and shoot tissue. The cellular regulatory mechanisms controlling the development ofplastids in young tissues are poorly understood, especially in the case of non-green plastids in roots. This investigation con-sequently aimed to elucidate the differences in plastid content, morphology and subcellular localization within epidermalcells derived from the root and shoot apical meristems (RAM and SAM respectively) of Arabidopsis thaliana.

Quantification of non-green plastids was facilitated via the use of confocal laser scanning microscopy in conjunctionwith the expression of plastid-targeted green fluorescent protein driven by a constitutive promoter. Characterization ofearly seedling development and tissue diversification was also achieved by assessing epidermal cell size relative to devel-opmental progression, ultimately facilitating comparative analyses of plastid dynamics on both a temporal and tissue-varietal basis.

The number of plastids in epidermal cells within RAM-derived tissue was shown to increase across regions of cell divisionbefore being regulated throughout subsequent zones of elongation and maturing root tissue. In contrast, epidermal cells ofthe hypocotyl exhibit a more generalized increase in plastid number and less strict maintenance of cell plan area coverageduring tissue expansion.

The findings presented here suggest the functioning of distinct mechanisms regulating plastid division and growth inrelation to cell size within shoot and root apical meristem-derived tissues.

Key words: Plastids, GFP, Epidermal Cells, Root, Hypocotyl, Arabidopsis thaliana

Submitted on 18 May 2016; editorial decision on 6 January 2017

†This research was carried out as part of my undergraduate dissertation, which attributed towards the degree of BSc (Hons) in Biotechnology atThe University of Nottingham, Sutton Bonington Campus, UK.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1© The Author 2017. Published by Oxford University Press.This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. Forcommercial re-use, please contact journals.permissions@oup.com

IntroductionPlastids are organelles that ubiquitously reside in the cells ofhigher plants, adopting different forms in order to facilitate adiverse array of intracellular functions. They are believed tohave originated through a primordial endosymbiotic eventwith a free-living cyanobacterium-like prokaryote and remainfundamentally distinct from other eukaryotic organelles(Gould, Waller and McFadden, 2008; Pyke, 2009). The fore-most role of plastids, or more specifically green-pigmentedchloroplasts, is the fixation of atmospheric carbon dioxideinto a plethora of organic molecules that facilitate growth anddevelopment (Robertson and Laetsch, 1974; Pyke, 2009).Unsurprisingly, due to the pivotal nature of photosynthesisamidst aims to improve its efficiency, chloroplasts withinfoliar tissue have been the focus of significantly higher levelsof research compared to plastids of other tissues. This is alsopartly due to methodological difficulties in quantifying non-green plastids before the advent of genetic manipulation, theuse of reporter gene constructs and confocal laser scanningmicroscopy (CLSM). These non-green plastids, however, areno less intrinsic to plant development.

Leucoplasts are one such plastid subgroup crucial to lipidbiosynthesis, amino acid metabolism and as storage sites forstarch grains (amyloplasts), lipids (elaioplasts) and proteins(aleuroplasts) (Ohlrogge and Browse, 1995; Finkemeier andLeister, 2010). Of these, amyloplasts are deemed particularlycritical to correct development as sources of long-term energyreserves essential for growth and the formation of viable seedsand desirable grain (MacCleery and Kiss, 1999; Pyke, 2009).They are also fundamental to plants’ gravitropism where suf-ficiently dense amyloplasts termed statoliths are translocatedby gravity and promote the redistribution of auxin throughmembrane-based influx/efflux proteins, enabling directionalgrowth under conditions of otherwise limited stimuli (Swarupet al., 2005).

In addition to key intracellular roles, plastids are alsoimportant for determining interactions between plants andtheir immediate environment. Chromoplasts, for example,synthesize and accumulate pigments including yellow xantho-phylls and red lycopene in tomatoes, which can encourage

pollen and seed dispersal by attracting pollinators or herbi-vores at specific developmental stages (Galpaz et al., 2006).Due to their fundamental roles in plant biology, the ability toaccumulate novel compounds and other unique advantages,plastids present significant targets for agronomic enhance-ment and new biotechnological applications. They havealready been widely used within the field of genetic manipula-tion (Medina-Bolíva and Cramer, 2004; Daniell, 2006;Clarke and Daniell, 2011) but a large potential for plastidgenetic engineering still exists, perhaps emerging fromincreased research into non-green plastids and their develop-mental biology.

Higher plant embryogenesis transforms a fertilized ovuminto a juvenile form of the plant lacking most species-specificfeatures (Fig. 1). Post-embryonic development subsequentlyoccurs from meristematic tissue in a progressive and highlyorganized wave-like manner, generating new tissue throughsuccessive cell replication and expansion (Malamy andBenfey, 1997; Cary, Che and Howell, 2002). The followingincrease in architectural complexity is thus heavily influencedby changing factors within the early stages of germination(Jürgens, 2001).

The total plastid content of all plant cells originates fromsmall, undifferentiated plastids termed proplastids foundwithin dividing cells of the root and shoot apical meristems(RAM and SAM, respectively) (Pyke, 2009; Finkemeier andLeister, 2010). They replicate via the prokaryotically-derivedmechanism of binary fission to form daughter plastids which,having segregated into individual cells during mitosis, differ-entiate between plastid varieties depending on intracellularconditions and gene expression (Pyke, 2009; Osteryoung andPyke, 2014). The renowned example of this is during chloro-plast development where, in the absence of light, proplastidsdevelop into etioplasts containing semi-crystalline prolamellarbodies which, when later exposed to light, transform into thecharacteristic, chlorophyll-containing thylakoid membranesof chloroplasts (Finkemeier and Leister, 2010). Whilst classi-fied into different plastid sub-types according to their functionand internal structure, differentiation between plastid types isboth highly prevalent and fundamental to their success asdynamic elements of plant biology.

Figure 1. Seed germination in A. thaliana. (A) Diagram of internalized embryo. (B) Germination observed from 0 to 72 h. Radicle emergence andtesta shedding were seen to occur at approximately 36 and 52 h after sowing, respectively.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research article Bioscience Horizons • Volume 10 2017

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2

Green fluorescent protein (GFP) has been used extensivelyin modified forms as a reporter for gene expression (Chalfieet al., 1994). Unlike other methods of analysing expression,its introduction within target gene constructs facilitates thedirect visualization of expression in vivo by naturally fluores-cing green when exposed to light in the blue to ultravioletrange (Chiu et al., 1996). Targeting GFP to plastids whenused in partnership with CLSM can therefore be used toquantify their size, structure and subcellular localization.

Whilst the mechanism of chloroplast division has beenstudied in depth (Maple and Møller, 2007; Osteryoung andPyke, 2014), relatively little is known about how other plastidvarieties divide, especially proplastids and other root-basedplastids. A wide array of plastid-derived structures includingstromules and vesicles have been observed in cultured root tis-sue however, suggesting that division mechanisms additionalto those of chloroplasts are present and required to facilitatethe regulation of a population of increasingly heterogeneousplastid forms (Pyke, 2013).

Considering the importance of plastids to plant biology andtheir potential in enhancing cultivars of high socioeconomicimportance, it is somewhat surprising that relatively limitedresearch has been undertaken into their development duringembryogenesis and germination; particularly into the differ-ences seen between the plastid dynamics of root and shootmeristem-derived tissues. As individual systems of gene expres-sion and development, these early differences could have pro-found impacts on general plant biology, crop improvementand biotechnology. Before further research can be performed,however, a detailed characterization of a suitable model systemis fundamental.

This investigation consequently aimed to elucidate some ofthe differences seen between meristem-derived tissues by visu-alizing plastid number and size across Arabidopsis thalianaseedling transects during the first 7 days after imbibition.

Materials and methodsSeeds of A. thaliana (var. Columbia) were donated by GeorgeBassell and originated from the study of Cutler et al. (2000).Through Agrobacterium-mediated transformation, a trans-gene containing plastid-targeted GFP under the control of aconstitutive promoter was introduced, facilitating the intra-cellular visualization of all plastid forms by GFP fluorescenceusing CLSM.

Growth mediaBacto-agar enriched with Murashige and Skoog (MS) saltswas used as an optimal growth medium that enabled unobtru-sive observation and excision of seedlings (Murashige andSkoog, 1962). All growth media was produced in the ratio of150ml purified water, 1.5 g sucrose and 0.645 g MS basalsalts mixed to a pH of 5.8–6.0. The resulting solution was

then added to 1.5 g bacto-agar and sterilized at 121°C for 1 husing an autoclave.

Seventy-five millilitre agar was poured into standard 120 ×120mm2 square petri dishes onto which two rows of 10 ster-ile seeds were evenly distributed.

Seed sterilization and growthWithin a laminar flow hood, seeds were suspended in a solu-tion of 0.75ml 50% bleach/water and 0.75ml 0.1% Triton/water. After 6 min, any superfluous solution was discardedfrom sedimented seeds before 1 ml 70% ethanol solution wasused to resuspend the seeds. Seed suspensions were theninverted over 20 s before excess solution was removed andseeds distributed on cooled agar.

Seeds were cooled to 4°C for 12 h in order to ensure evengermination before being placed in a growth room main-tained at 20–25°C on a 24 h light cycle. A total of twentyplates containing 400 seeds were monitored over 156 h in12-h increments for comparative growth between root andhypocotyl tissue whilst, where appropriate, seedlings wereexcised for further analysis.

Confocal microscopy and image analysisTen seedlings were harvested every 12 h from 60 h after plat-ing (until 156 h) and were imaged using a Leica SP2 confocallaser scanning microscope (Leica Microsystems, Heidelberg,Germany). A single plane of peak GFP and chlorophyll fluor-escence within the epidermal layer was measured in each sam-ple from the root tip, across the root–hypocotyl tissueboundary and into hypocotyl tissue, retaining accurate plastidquantification throughout all tissues under investigation.

As an indication of whether the observed prevalence ofplastids was representative of each cell’s actual content, Zstacks were performed and assessed on samples of 20 similarsized epidermal cells in each major tissue variety and a rangeof seedlings (4 × 96 h, 120 h, 144 h and 156 h). GFP imagingwas accomplished by directing a 488 nm laser for excitationonto samples while detecting fluorescence at an emissionwavelength of 509 nm. Chlorophyll autofluorescence wasimaged at a peak emission wavelength of 673 nm whilstbright field images of tissue were also collected.

Cell walls were labelled with propidium iodide (PI); seed-ling tissue samples were excised using a scalpel and soakedfor 5 min in 3ml 0.5 μm PI solution before being carefullyrinsed several times with water and imaged (excitation wave-length = 535 nm, emission wavelength = 617 nm).

All recorded images consisted of an average of eight com-plete scans that were later compiled into single, layeredimages of each channel across the entire length of individualseedlings (Figs 3, 4B, 5D and 7B). These images were thenanalysed using ImageJ software (http://imagej.nih.gov/ij/download.html, downloaded on 20/10/2014), ensuring atleast 80% of visible epidermal cells were assessed for plan cell

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Bioscience Horizons • Volume 10 2017 Research article

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3

area, plastid number and percentage coverage (percentageplan cell area occupied by plastids). Statistical analyses andgraphs were performed and produced respectively usingMicrosoft Excel 2013 (Figs 2, 4A, 6, 7A and 8).

Results

Seedling growth characterizationMeasuring the average length of the hypocotyl and root tissueduring germination showed that root growth contributed

more to total seedling length than hypocotyl growth (Fig. 2).Total seedling, root and hypocotyl tissue length were all alsoseen to increase by a near-consistent 3% per hour after rootemergence, suggesting a conserved and relatively stable rateof growth within the two types of tissue.

Tissue under investigation can be classified into regionsbased on gross tissue morphology, cell replication and size.Moving across root tissue from the root tip to the root–hypocotyl tissue boundary, successive zones of high (butdecreasing) cell division, elongation and root hair tissue exist.Many of these tissues were seen to overlap however, restrict-ing the ability to conclusively investigate discrete tissue var-ieties. Within the hypocotyl, less visually recognizablechanges in morphology were observed during the early stagesof development except for a notable increase in chlorophyllcontent revealed by autofluorescence (Fig. 3). Regions ofhigher rates of cell division, elongation and mature hypocotyltissue were also present and can be qualified by changes inepidermal cell size (Fig. 4A).

An increase in epidermal cell size with distance from roottip was seen to occur consistently in different seedlings atnear-constant distances (Fig. 4A). In root tissue, the zone ofcell division extended approximately 1100 μm from the roottip, within which cells of the root cap were shown to uni-formly increase in plan cell area from an average of 85.7 to351.2 μm2. Subsequently, a major increase in cell size wasobserved beyond 1100 μm where average epidermal cell sizeincreased by roughly 70 μm2 per 100 μm until a distance of

Figure 3. Superimposed confocal image of chlorophyll fluorescence(red) and bright field image of A. thaliana seedling after 52 h. Redchlorophyll fluorescence can be seen in the hypocotyl and cotyledonsbut not within root tissue after the root–hypocotyl tissue boundary.

Figure 4. (A) Exemplified trend within 96 h A. thaliana of changingepidermal plan cell area against distance from root tip. Root epidermalcells ( ), hypocotyl epidermal cells ( ). (B) Annotated compilation ofconfocal bright field channel images of 96 h A. thaliana seedling acrossthe root–hypocotyl transect. Labels indicate the approximate regionsof distinct tissue morphology.

Figure 2. Length of total A. thaliana seedling ( ), root ( ) andhypocotyl ( ) tissue during germination. Error bars represent standarderror. Number of seedlings assessed = 340.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research article Bioscience Horizons • Volume 10 2017

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4

4500 μm from the root tip. After 4500 μm until the root–hypocotyl tissue boundary, rate of cell expansion in all sam-ples declined and epidermal cells averaged a plan area of3497 μm2.

Epidermal cells within the hypocotyl followed a similartrend of progressive expansion as distance from the SAMincreased but exhibited higher levels of variation (Fig. 4A).Within the zone of elongation, cells were seen to consistentlyincrease at an average rate of 142 μm2 per 100 μm; doublethat seen in comparable root tissue. Hypocotyl epidermalcell size was also seen to stabilize at comparatively lowerlevels than root epidermal cells on the root–hypocotyl tissueboundary; on average approximately 2100 and 3600 μm2,respectively.

These changes to epidermal cell size were seen to bedependent on the morphology and developmental status ofhypocotyl and root tissues from 60 to 156 h in all samples.Epidermal cell size was consequently used as a way to qualifydevelopmental progression against the distinct regions ofplastid prevalence seen across the root–hypocotyl boundary(Fig. 5A), tissue within the zone of elongation/mature root(Fig. 5B) and in the root tip (Fig. 5C).

Plastid number with epidermal cell sizeThe number of plastids within root epidermal cells exhibitedtwo principal stages when plotted against cell plan area(Fig. 6A). Within cells of an average size of 85.7 μm2 upwardsto 500 μm2, an increase in plastid number was seen from 3 to16. Past the zone of cell replication characterized by plan cellareas higher than 500 μm2, a relatively consistent number ofplastids were subsequently observed averaging slightly higherthan 17 (17.44) but with a range of 9–26 (SD = 4.18). Inhypocotyl tissue, the number of plastids per epidermal cellremained within the range of 3–10 across cell sizes below500 μm2 (average of 5.84, SD = 1.95) (Fig. 6B). Plastid num-ber subsequently increased to approximately 26 (26.44)within epidermal cells larger than 1700 μm2. Variation inplastid number was seen however, consistently ranging from20 to 36 plastids per cell (SD = 4.62).

Plastid coverage against transect distanceand epidermal cell sizeAcross all samples, epidermal plan cell area plastid coveragein root tissue was seen to decrease from an average of 39.7%within the root tip and across the zone of cell division (up to1100 μm from the root tip) to 20.8% throughout the zones ofcell elongation and mature root tissue (Fig. 7A). In hypocotylepidermal cells, plastid coverage followed a similar trend. Asdistance from the SAM increased, percentage plastid coveragewas seen to drop from an average of 52.1% within the zoneof cell division (up to 900 μm from SAM) to 24.9% through-out the remaining tissue. In all equivalent hypocotyl and rootepidermal cells, plastid coverage was seen to be generallyhigher in hypocotyl tissue.

In root tissue, percentage plastid coverage was observed tobe higher within epidermal cells of a plan area lower than500 μm2 (Fig. 8A). Due to their size however, relatively smallchanges caused large differences leading to variation in plastidcoverage from 25% to 55% (average = 37.7%, SD = 8.8%).In cells indicative of the zone of elongation and mature roottissue (cell sizes of 500–4500 μm2), percentage coverage wasshown to average 20.2% (SD = 5.1%) (Fig. 8A).

By comparison, hypocotyl tissue exhibited a less fixedtrend of reducing plastid coverage against increasing epider-mal cell size (Fig. 8B). Higher levels of variation from 20%to 75% were seen across all cells lower than 2000 μm2

(average = 41.8%, SD = 13.6%). In epidermal cells largerthan 2000 μm2, coverage was seen to decrease until an aver-age of 28.3% (SD = 8.5%).

In order to qualify if the recorded number of plastids andpercentage plan cell area coverage correlated to the actualcontent of epidermal cells, Z stacks of individual cells were

Figure 5. Superimposed confocal GFP and bright field images of 96 hA. thaliana seedling showing the changing plastid dynamics withinepidermal cells of: (A) the root–hypocotyl tissue boundary; (B) matureroot tissue and (C) root tip. (D) Compilation of superimposed confocalGFP and bright field channel images of whole 96 h A. thaliana seedlingacross the root–hypocotyl tissue transect. Labels indicate theapproximate regions of distinct tissue morphology and root–hypocotyltissue boundary.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Bioscience Horizons • Volume 10 2017 Research article

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5

performed. Due to difficulties in determining intercellularboundaries on a Z-plane and reduced fluorescence however,no conclusive scale factor between observable plastid numberand total plastid content could be established across the rangeof tissues investigated. A statistically significant correlationbetween increases in observed plastid content and that seenwithin Z stacks was found though, suggesting that data col-lected can be used for comparative analyses. Intracellularplastid numbers measured also corresponded generally to pre-vious studies; 10–20 proplastids within meristematic cells andup to 100 chloroplasts depending on cell size and function(Pyke, 1999; Segui-Simarro and Staehelin, 2009).

Plastid morphology and subcellularlocalizationPost-root tissue containing statoliths but within the zone ofcell replication, no significant patterning of intracellular local-ization was observed in any samples. Plastids were seen to bedistributed between cell peripheries, cytoplasm and in closeproximity to nuclei (Fig. 9A). Thin, stromule-like projectionsand ‘dumbbell’-shaped plastids were also seen within root tiptissue and regions of cell elongation (Fig. 9B) but not as preva-lently in zones of cell division and hypocotyl cells (data notshown). Within cells of the zone of elongation and matureroot tissue, plastid morphology was seen to stabilize intomostly spherical and elliptical forms distributed along inter-cellular boundaries (Fig. 9C).

Figure 7. (A) Exemplified trend within 96 h A. thaliana of changingpercentage epidermal cell plastid coverage against distance from roottip. Root epidermal cells ( ), hypocotyl epidermal cells ( ). (B)Annotated compilation of superimposed confocal GFP and bright fieldchannel images of 96 h A. thaliana seedling across the root–hypocotyltissue transect. Labels indicate the approximate regions of distincttissue morphology.

Figure 6. Aggregated 96 h, 120 h and 144 h A. thaliana seedling dataof changing number of plastids per cell against epidermal plan cellarea. (A) Root epidermal cells ( ). (B) Hypocotyl epidermal cells ( ).

Figure 8. Aggregated 96 h, 120 h and 144 h A. thaliana seedling dataof changing percentage epidermal cell plastid coverage againstepidermal plan cell area. (A) Root epidermal cells ( ). (B) Hypocotylepidermal cells ( ).

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research article Bioscience Horizons • Volume 10 2017

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6

Throughout the zones of cell replication and elongation ofthe hypocotyl, larger plastids were observed compared towithin the equivalent root epidermal layer and, whilstdumbbell-shaped plastids and stromules were seen towardsthe SAM, their prevalence was significantly lower comparedto RAM-derived tissue. In all instances, plastids of the hypo-cotyl were also predominantly localized around cell peripher-ies (Fig. 9D).

Discussion

Interpretation of resultsPlastid dynamics within hypocotyl and RAM-derived epider-mal cells of A. thaliana have been shown to deviate in terms ofplastid number, plan cell area coverage and morphology atnear-fixed stages of early plant development. Characterizationof A. thaliana growth and epidermal cell plan area assisted inqualifying these changes whilst additionally providing a meth-od of assessing plastid dynamics across a range of sample seed-lings and tissue varieties (Fig. 4).

In root meristematic tissue and the zone of cell division, theobservable number of plastids per epidermal cell was seen toincrease with cell size, and thus distance from meristematic

tissue, whilst a relatively stable area of plastid cell coveragewas maintained (Figs 6A and 8A, respectively). By compari-son, the number of plastids within equivalent tissues of thehypocotyl remained within the lower spectrum of those seenin root tissue whilst a higher discrepancy for area of plastidcell coverage was observed (Figs 6B and 8B, respectively).

This suggests that within actively dividing tissue, plastidsare also replicating whilst a system for regulating plastid num-ber and size is in effect. It also implies that within hypocotylregions of cell replication, somewhat less rigorous or func-tionally different regulatory pathways exist.

Within maturing root tissue, percentage plastid coveragewas seen to be approximately half that of replicating regions(Fig. 8A) whilst plastid number was also relatively tightly main-tained (Fig. 6A). This suggests the presence of strong regulatorysystems assessing both intracellular plastid number and size inroot epidermal cells. Conversely, epidermal cells of the hypo-cotyl exhibited an increase in plastid number throughout thezone of elongation and mature tissue alongside a reduction inplastid coverage (Figs 6B and 8B, respectively). This indicatesthat plastid number rather than size is increasing whilst allud-ing to additional differences in regulatory systems between thetwo tissues. The disparities in plastid replication rates andmorphology may have a number of evolutionary advantages,potentially including the preservation of resources by inhibitingthe formation of functionally superfluous plastids.

Within the hypocotyl and subsequent foliar tissue, highernumbers of plastids (specifically chloroplasts) are beneficialdue to their increased presence enabling more efficient lightabsorbance and photosynthesis (Finkemeier and Leister,2010). Increasingly higher levels of peripheral localizationand chlorophyll autofluorescence were seen across transectsof the hypocotyl, signifying a definitive increase in chloroplastdifferentiation and functional relocation (Figs 3 and 9D)(Trojan and Gabrys, 1996). It is unclear, however, what fac-tors induce these changes and limit potential plastid differenti-ation. All seedlings were grown laterally across agar in a 24-hlight cycle, suggesting that any abiotic factors present equallyaffected the two tissues, potentially implying that some degreeof plastid predefinition according to function may exist.

The presence of stromules (stroma-filled tubules) projectingfrom plastids (Fig. 9B) have been shown to facilitate the traffick-ing of molecules between plastids (Hanson and Sattarzadeh,2011), however, their role in the exchange of molecules remainsa controversial subject. Their presence could still be potentiallyimportant within the zone of elongation and mature root tissuewhere root hairs are present and nutrient uptake occurs(Benning, Xu and Awai, 2006). Plastids have also been shownto play a major role in nitrate assimilation by reducing it intonitrite (formed within the cytosol from nitrate by nitrate reduc-tase) and further into ammonia using nitrite reductase(Finkemeier and Leister, 2010).

Increased plastid surface area and molecular traffickingfacilitated by stromules may contribute to more efficient

Figure 9. Superimposed confocal GFP and bright field channelimages showing subcellular plastid localization within 96 h (A and B)and 120 h (C and D) A. thaliana epidermal cells. (A) Irregular patterningwithin replicating root cells. (B) Stromules and dividing (dumbbell-shaped) plastids within elongating root cells highlighted by broad andthin arrowheads respectively. (C) Peripheral patterning withinelongating/mature root cells. Dashed lines indicate approximatelocation of plant cell walls (D) Peripheral patterning within maturehypocotyl tissue.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Bioscience Horizons • Volume 10 2017 Research article

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7

nutrient absorption, and is perhaps why the presence of plas-tids within RAM-derived tissues is so highly regulated; espe-cially during early development where the use of nutrientsmust be optimized for survival. It is also interesting to notethe discrepancy seen in equivalent hypocotyl tissue wherestromules were seen less frequently, both here and in previousinvestigations (Pyke and Howells, 2002).

Plastid replication and regulatorymechanismsExtensive research has been undertaken into plastid divisionand regulation by characterizing Arabidopsis accumulationand replication of chloroplasts (arc) mutants. In particular,arc6, a mutant known to interfere with proplastid division inroot and shoot meristematic tissues has helped define distinctgene regions fundamental to controlling intracellular plastidnumber (Robertson, Pyke and Leech, 1995; Vitha et al.,2003).

Plastid division is facilitated through the formation of adouble ringed protein complex encircling dividing plastids attheir centre, perpendicular to their longitudinal axis. One ringforms on the stromal side of plastids’ inner envelope whilstthe other forms on the cytosolic side of their outer envelope(Pyke, 1999). Using a dynamin-related protein that producescytokinetic force (DRP5B/ARC5), this ring constricts untilenvelope membranes deliquesce and reform to produce twoplastids (Maple andMøller, 2007; Yang et al., 2008).

Increased numbers of dumbbell-shaped plastids are there-fore representative of higher levels of plastid replication, asseen within cells of both RAM (Fig. 9B) and SAM prior to thezone of division and in later, maturing tissue (data notshown). Whilst morphology of plastids can indicate levels ofreplication, the disparities seen in the regulation of plastidcontent between meristematic and mature root tissues are per-haps derived from a combination of plastids’ primordial ori-gins and homoeostatic mechanisms.

During the simultaneous evolution of the ancestral precur-sors to plastids and their eukaryotic hosts, genes once presentin the endosymbiont’s genome were either lost, retained ortranslocated to the host’s nucleus and levels of synchronybetween themwere established (Gould et al., 2008). Of the pro-teins encoded by such genes, some are essential to plastid repli-cation and have been shown to accumulate during discretephases of the cell cycle in accordance with cyclin-dependentkinases (CDKs). These include ARC3, MinD, MinE (specifythe site of plastid division), FtsZ proteins (tubulin-like proteinsthat form the early ring structure), ARC6 (recruits other pro-teins to the ring/site of division) and DRP5B/ARC5 (generatesthe cytokinetic force necessary for division) (Okazaki, Kabeyaand Miyagishima, 2010). Other essential proteins, such as thenuclear-encoded plastid division proteins (PDVs), have beenshown to regulate DRP5B/ARC5 and consequently plastid div-ision according to cell cycle progression and/or differentiationdetermined by cytokinin, a phytohormone fundamental to

shoot and root tissue formation in early embryogenesis(Okazaki et al., 2010; Holtsmark et al., 2013). The phase ofindividual cells’ replicative cycles, tightly coupled nuclear-based gene expression and tissue variety have the potential todirectly mediate plastid division (Gillard et al., 2008).

Homeostatic mechanisms also control the levels of intra-cellular PDVs and thus rates of plastid division. One suchmechanism is the ubiquitin-proteasome system formed of anumber of protein complexes capable of degrading unneededor damaged target proteins by proteolysis (proteasomes) anda regulatory protein that facilitates the post-translational ‘tag-ging’ of target proteins for degradation (ubiquitin) (Haaset al., 1982; Kimura and Tanaka, 2010). Through targetedubiquitin-activating enzymes, PDVs can thus be degraded andplastid division regulated in accordance to the expression ofspecific genes includingMinC; an inhibitor of FtsZ’s polymer-ization into the contractile ring’s precursor (Margolin, 2005).

Coupling of plastid and cellular divisionWhilst many of these regulatory mechanisms are shown tocoincide with discrete phases of the cell cycle, they cannot beconsidered to be ‘coupled’ with the division of plastids, as eachprocess can be altered independently without impairing theother (Osteryoung and Nunnari, 2003). In arc mutants andplants overexpressing PDVs (a factor that counterintuitivelyinhibits plastid division), no defects in development or cell div-ision have been seen (Pyke and Leech, 1992; Osteryoung et al.,1998). Reciprocally, stopping cell division via inhibition ofCDKs does not, whilst seriously hindering normal plant devel-opment, necessarily affect plastid division as seen in plantsoverexpressing NtKIS1a; a highly characterized CDK inhibitor(Jasinski et al., 2002, 2003).

Future research and potentials of definedplastid dynamicsFurther research into the regulatory factors differing betweenearly hypocotyl and root tissue could be undertaken by profil-ing gene expression whilst assessing the plastid dynamics ofrespective tissues (Bustin, 2002; Trevino, Falciani andBarrera-Saldaña, 2007). The extension of this study into thecell layers beyond the epidermis will also be a necessary focusfor defining plastid dynamics against plant developmentbecause, whilst epidermal cells are considerably easier toassess and provide comparative analysis across both temporaland surface tissue-based development, they do not form cer-tain key tissues of mature plants. Plastid prevalence could alsobe compared between wild type A. thaliana, meristematicmutants such as those investigated by Aida, Ishida andTasaka (1999) and important crop cultivars in order to qual-ify the impacts of early plastid variation upon later growth.

Ultimately, the data collected here highlights the paucity ofknowledge relating to the early discrepancies of plastiddynamics within early plant tissues and, whether due tochanges in gene expression, cellular identity and/or abiotic

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research article Bioscience Horizons • Volume 10 2017

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8

conditions, these observations could correlate to plastidprevalence within mature plant tissue. Defining the causes ofearly plastid dynamics could be potentially exploited forfuture research alongside the development of biofactories ofimportant molecules and enhanced cultivars of high socio-economic importance.

Author biographyMolecular biology appeals to his avid interest in science, bothin terms of how elements intrinsic to life function and how theycan be adapted. He aims to forge a career in science under-pinned by this fascination of how minute intracellular changesinduce huge phenotypic impacts, specifically researchingmolecular plant pathology and crop cultivar improvementthrough genetic engineering. In 2015, he graduated from TheUniversity of Nottingham with a first-class honours degree inBiotechnology, and is about to embark on a PhD at TheUniversity of Warwick where he will research resistance toTurnip mosaic virus in members of the Brassica genus.

AcknowledgementsI would like to especially thank Dr Kevin Pyke for his signifi-cant guidance throughout this study and all the fellow under-graduate students, postgraduate students and staff of UoNSutton Bonington campus’ plant science laboratories. Addi-tional thanks to George Bassell for donating the transgenic A.thaliana seeds used throughout this experiment and the grant-aided support that research at The University of Nottinghamreceives from the Biotechnological and Biological SciencesResearch Council UK (BBSRC).

ReferencesAida, M., Ishida, T. and Tasaka, M. (1999) Shoot apical meristem and

cotyledon formation during Arabidopsis embryogenesis: interactionamong the CUP-SHAPED COTYLEDON and SHOOT MERISTEMLESSgenes, Development, 126 (8), 1563–1570.

Benning, C., Xu, C. and Awai, K. (2006) Non-vesicular and vesicular lipidtrafficking involving plastids, Current Opinion in Biotechnology, 9,241–247.

Bustin, S. A. (2002) Quantification of mRNA using real-time reverse tran-scription PCR (RT-PCR): trends and problems, Journal of MolecularEndocrinology, 29, 23–29.

Cary, A. J., Che, P. and Howell, S. H. (2002) Developmental events andshoot apical meristem gene expression patterns during shoot devel-opment in Arabidopsis thaliana, The Plant Journal, 32, 867–877.

Chalfie, M., Tu, Y., Euskirchen, G. et al. (1994) Green fluorescent proteinas a marker for gene expression, Science, 263, 802–805.

Chiu, W. L., Niwa, Y., Zeng, W. et al. (1996) Engineered GFP as a vitalreporter in plants, Current Biology, 6 (3), 325–330.

Clarke, J. L. and Daniell, H. (2011) Plastid biotechnology for crop pro-duction: present status and future perspectives, Plant MolecularBiology, 76 (3–5), 211–220.

Cutler, S. R., Ehrhardt, D. W., Griffitts, J. S. et al. (2000) Random GFP:cDNA fusions enable visualization of subcellular structures in cellsof Arabidopsis at a high frequency, Proceedings of the NationalAcademy of Sciences of the USA, 97 (7), 3718–3723.

Daniell, H. (2006) Production of biopharmaceuticals and vaccines inplants via the chloroplast genome, Biotechnology Journal, 1 (10),1071–1079.

Finkemeier, I. and Leister, D. (2010) Plant chloroplasts and other plas-tids, in Encyclopedia of Life Sciences (ELS), John Wiley and Sons, Ltd,Chichester, United Kingdom.

Galpaz, N., Ronen, G., Khalfa, Z. et al. (2006) A chromoplast-specific carotenoid biosynthesis pathway is revealed by clon-ing of the tomato white-flower locus, The Plant Cell, 18 (8),1947–1960.

Gillard, J., Devos, V., Huysman, M. J. J. et al. (2008) Physiological andtranscriptomic evidence for a close coupling between chloroplastontogeny and cell cycle progression in the pennate diatomSeminavis robusta, Plant Physiology, 148 (3), 1394–1411.

Gould, S. B., Waller, R. F. and McFadden, G. I. (2008) Plastid evolution,Annual Review of Plant Biology, 59, 491–517.

Haas, A. L., Warms, J. V., Hershko, A. et al. (1982) Ubiquitin-activatingenzyme: mechanism and role in protein-ubiquitin conjugation, TheJournal of Biological Chemistry, 25 (5), 2543–2548.

Hanson, M. R. and Sattarzadeh, A. (2011) Stromules: recent insights intoa long neglected feature of plastid morphology and function, PlantPhysiology, 155, 1486–1492.

Holtsmark, I., Lee, S., Lunde, K. A. et al. (2013) Plastid division control:the PDV proteins regulate DRP5B dynamin activity, Plant MolecularBiology, 82 (3), 255–266.

Jasinski, S., Leite, C. S., Domenichini, S. et al. (2003) NtKIS2, a noveltobacco cyclin-dependent kinase inhibitor is differentiallyexpressed during the cell-cycle and plant development, PlantPhysiology and Biochemistry, 41, 667–676.

Jasinski, S., Riou-Khamlichi, C., Roche, O. et al. (2002) The CDK inhibitorNtKIS1a is involved in plant development, endoreduplication andrestores normal development of cyclin D3; 1-overexpressing plants,Journal of Cell Sciences, 115 (5), 973–982.

Jürgens, G. (2001) Apical-basal pattern formation in Arabidopsisembryogenesis, The European Molecular Biology Organization, 20(14), 3609–3616.

Kimura, T. and Tanaka, K. (2010) Regulatory mechanisms in the controlof ubiquitin homeostasis, The Journal of Biochemistry, 147 (6),793–798.

MacCleery, S. A. and Kiss, J. Z. (1999) Plastid sedimentation kinetics inroots of wild-type and starch deficient mutants of Arabidopsis, PlantPhysiology, 120, 183–192.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Bioscience Horizons • Volume 10 2017 Research article

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9

Malamy, J. E. and Benfey, P. N. (1997) Organization and cell differentiationin lateral roots of Arabidopsis thaliana, Development, 124, 33–44.

Maple, J. and Møller, S. G. (2007) Plastid division: evolution mechanismand complexity, Annals of Botany, 99, 565–579.

Margolin, W. (2005) FtsZ and the division of prokaryotic cells and orga-nelles, Nature Reviews Molecular Cell Biology, 6, 862–871.

Medina-Bolíva, F. and Cramer, C. (2004) Production of recombinantproteins by hairy roots cultured in plastic sleeve bioreactors,Methods in Molecular Biology, 267, 351–363.

Murashige, T. and Skoog, F. (1962) A revised medium for rapid growthand bioassays with tobacco tissue cultures, Physiologica Plantarum,15 (3), 473–497.

Ohlrogge, J. and Browse, J. (1995) Lipid biosynthesis, Plant Cell, 7, 957–970.

Okazaki, K., Kabeya, Y. and Miyagishima, S. Y. (2010) The evolution ofthe regulatory mechanism of chloroplast division, Plant Signallingand Behavior, 5 (2), 164–167.

Osteryoung, K. W. and Nunnari, J. (2003) The division of endosymbioticorganelles, Science, 302, 1698–1704.

Osteryoung, K. W. and Pyke, K. A. (2014) Division and dynamic morph-ology of plastids, Annual Review of Plant Biology, 65, 443–472.

Osteryoung, K. W., Stokes, K. D., Rutherford, S. M. et al. (1998)Chloroplast division in higher plants requires members of two func-tionally divergent gene families with homology to bacterial ftsZ,The Plant Cell, 10, 1991–2004.

Pyke, K. A. (1999) Plastid division and development, The Plant Cell, 11(4), 549–556.

Pyke, K. A. (2009) Plastid Biology, Cambridge University Press, UnitedKingdom.

Pyke, K. A. (2013) Divide and shape: an edosymbiont in action, Planta,237, 381–387.

Pyke, K. A. and Howells, C. A. (2002) Plastid and stromule morphogen-esis in tomato, Annals of Botany, 90, 559–566.

Pyke, K. A. and Leech, R. (1992) Chloroplast division and expansion isradically altered by nuclear mutations in Arabidopsis thaliana, PlantPhysiology, 99, 1005–1008.

Robertson, D. and Laetsch, W. M. (1974) Structure and function ofdeveloping barley plastids, Plant Physiology, 54 (2), 148–159.

Robertson, E. J., Pyke, K. A. and Leech, R. M. (1995) arc6, an extremechloroplast division mutant of Arabidopsis also alters proplastidproliferation and morphology in shoot and root apices, Journal ofCell Science, 108, 2937–2944.

Segui-Simarro, J. M. and Staehelin, L. A. (2009) Mitochondrial reticula-tion in shoot apical meristem cells of Arabidopsis provides a mech-anism for homogenization of mtDNA prior to gamete formation,Plant Signalling and Behavior, 4, 168–171.

Swarup, R., Kramer, E. M., Perry, P. et al. (2005) Root gravitropismrequires lateral root cap and epidermal cells for transport andresponse to a mobile auxin signal, Nature Cell Biology, 7 (11),1057–1065.

Trevino, V., Falciani, F. and Barrera-Saldaña, H. A. (2007) DNAMicroarrays: a powerful genomic tool for biomedical and clinicalresearch,Molecular Medicine, 13 (9–10), 527–541.

Trojan, A. and Gabrys, H. (1996) Chloroplast distribution in Arabidopsisthaliana (L.) depends on light conditions during growth, PlantPhysiology, 111, 419–425.

Vitha, S., Froehlich, J. E., Koksharova, O. et al. (2003) ARC6 is a J-domainplastid division protein and an evolutionary descendent of thecyanobacterial cell division process, Journal of Bacteriology, 15,1918–1933.

Yang, Y., Glynn, J. M., Bradley, J. S. C. et al. (2008) Plastid division: acrosstime and space, Current Opinion in Plant Biology, 11, 577–584.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research article Bioscience Horizons • Volume 10 2017

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10