RESEARCH ARTICLE Principles underlying chromatophore ... · Accepted 1 June 2011 SUMMARY The goal...

3423

INTRODUCTIONCamouflage plays a key role in the lives of many animals both asa method of defense enabling prey to hide from predators and as ahunting strategy allowing predators to stalk prey without detection(Stevens and Merilaita, 2009). One important aspect of evadingdetection by predators is the use of body patterning, in which coloredpatterns are displayed by an organism on its exterior (Cott, 1940).Although there are countless numbers of different camouflage bodypatterns to help animals hide in their natural environments, oftenthese patterns can be categorized into three basic classifications:uniform, mottle and disruptive (Cott, 1940; Hanlon and Messenger,1988; Barbosa et al., 2007). The vast majority of animals utilizeonly one static body pattern expressed throughout their livesregardless of environmental changes (Carvalho et al., 2006; Herbert,1974; Johnsen, 2001; Sazima et al., 2006; Fox and Vevers, 1960).Some species can slowly change their coloration to reflect changesin the environment, e.g. the snowshoe hare, which alters its fur colorfrom rusty brown in the summer to solid white in the winter(Severaid, 1945). Only a small percentage of animals, however, canactively change their coloration in real-time to reflect variations intheir background.

Coleoid cephalopods, unshelled mollusks including cuttlefish,squid and octopus, are among the very few organisms capable ofchanging their skin color and texture to match their background(Boycott, 1961; Hanlon and Messenger, 1996; Messenger, 2001).The control of body patterning is so rapid – less than 1s – that these

organisms are capable of producing dynamic body patterns such asthe ‘passing wave’, which mimics the movement of waves on theocean floor (Hanlon and Messenger, 1988).

Cephalopods achieve sophisticated control over their bodypatterning in large part because of the presence in their skin of threedifferent classes of specialized coloration cells known as iridiphores,leucophores and chromatophores (Brocco and Cloney, 1980; Packard,1985; Hanlon and Messenger, 1988). Of these, the chromatophoresare perhaps the most interesting because of their unique ability torapidly change their shape through a specialized neuromuscularcontrol system. Chromatophore cells have several, unique cellularspecializations. Each contains one of three classes of cytoplasmicallylocalized pigment molecules: black, reddish-brown or yellow.Attached to the highly infolded and very elastic plasma membraneare 6–20 radially arrayed chromatophore muscles directly innervatedby the central nervous system (Cloney and Florey, 1968).Simultaneous contraction of all chromatophore muscles results in therapid expansion of the pigmented chromatophore cell up to 10 timesits original diameter, from ~30m when condensed to ~300m whenfully expanded (Cloney and Florey, 1968; Loi and Tublitz, 1998). Asingle chromatophore cell, its surrounding chromatophore musclesand adjacent support cells are collectively referred to as achromatophore organ (Cloney and Florey, 1968).

While all unshelled cephalopods have chromatophore cells andchromatophore organs, their number, distribution and function varyduring post-embryonic maturation and across species (Packard and

The Journal of Experimental Biology 214, 3423-3432© 2011. Published by The Company of Biologists Ltddoi:10.1242/jeb.055251

RESEARCH ARTICLE

Principles underlying chromatophore addition during maturation in the Europeancuttlefish, Sepia officinalis

Jarred Yacob, Alexandra Cosima Lewis, Allyson Gosling, Debra H. J. St Hilaire, Lindsay Tesar, Michelle McRae and Nathan J. Tublitz*

Institute of Neuroscience, University of Oregon, Eugene, OR 97403, USA*Author for correspondence ([email protected])

Accepted 1 June 2011

SUMMARYThe goal of this work was to identify some of the principles underlying chromatophore growth and development in the Europeancuttlefish, Sepia officinalis. One set of experiments used a regeneration model to follow the re-growth of black chromatophoresfor 30 days following excision of a small piece of fin tissue. A separate set of experiments tracked and analyzed the addition ofnew fin chromatophores during a month of normal, undisturbed growth. We also followed the development of individualchromatophores from their initial appearance to full maturation to determine whether their color type was fixed. Based on theresults of these studies, we propose five guiding principles for chromatophore growth and maturation. (1) The threechromatophore cell types – black, reddish-brown and yellow – are present at different spatial frequencies in the cuttlefish fin. (2)During normal growth, new chromatophores are inserted at a higher spatial frequency than existing (control) chromatophores ofthe same color type. (3) In regenerating tissue, new black chromatophores are initially added at low spatial frequencies. Asregeneration continues, new black chromatophores appear at increasing spatial frequencies until they are inserted at a spatialfrequency higher than that observed in control tissue, similar to the way in which chromatophores were observed to be added innormally growing tissue. (4) All chromatophores first appear as pale orange cells and slowly darken into their respective colortypes without passing through intermediate color stages. (5) New black chromatophores undergo a doubling in size as theymature, while reddish-brown and yellow chromatophores do not grow at all after they are inserted in the dermis.

Key words: Sepia officinalis, cephalopod, body patterning, camouflage, chromatophore, cuttlefish.

THE JOURNAL OF EXPERIMENTAL BIOLOGY

3424

Sanders, 1969; Packard, 1985; Hanlon and Messenger, 1988;Shohet et al., 2007). At birth, a cuttlefish hatchling is generally lessthan a few centimeters in size and contains only several hundredchromatophores on its entire body (Hanley et al., 1998; Domingues,2001). As the individual grows, new chromatophores are added untileach adult has hundreds of thousands of chromatophores spreadacross its body in a precise array, with each color class ofchromatophore lying in its own epidermal layer (Cloney and Florey,1968; Packard and Hochberg, 1977; Cloney and Brocco, 1983;Packard, 1985; Hanlon and Messenger, 1996). Regular location andspacing of chromatophores is essential for accurate and functionalbody patterning. Any gap or disruption in the chromatophore arraycould substantially compromise the cuttlefish’s ability to camouflageproperly, increasing the risk of predation and making it difficult tocapture prey (Hanlon, 2007). For this reason it is crucial for newchromatophores to be added in a non-random manner to maintainconsistent chromatophore spacing necessary to generate seamlessbody patterns.

The purpose of this study was to determine the general principlesunderlying the growth and addition of new chromatophore cells. Everynew chromatophore is seamlessly inserted and functionally integratedinto the existing chromatophore matrix within a few days afterappearance without hindering the ability of the organism to producebody patterns. The addition of so many new, functionalchromatophores into a pre-existing chromatophore system raisesmultiple developmental and neural control issues including theprecision of their insertion location and the ability of the nervoussystem to precisely wire up new chromatophores into the existingarray. We focused on the former issue as a foundation for a futureinvestigation into the latter. This study used the European cuttlefish,Sepia officinalis, because it is easily reared in the laboratory andbecause of the wealth of literature on its body coloration patterns.Chromatophore addition was analyzed in normal growth andregeneration because each provided different experimental conditionsto explore and understand the principles underlying this interestingphenomenon. The results of this study should enhance ourunderstanding of the mechanisms by which cephalopods achieveadaptive body patterning for camouflage and communication.

MATERIALS AND METHODSAnimal subjects

Fourteen European cuttlefish, S. officinalis L. (age at beginning ofstudy: mean 3months, range 2weeks to 7months; mean mantlelength 7.5cm; range 3.2–9.1cm) were used for filming andobservation of fin growth. All test subjects were purchased fromthe National Resource Center for Cephalopods, Marine BiomedicalInstitute, Galveston, TX, USA, and were housed in two, 473lartificial seawater (ASW) tanks (122�60�60cm) at the Instituteof Neuroscience at the University of Oregon. Water conditions andculturing practices were identical to those previously described (Loiand Tublitz, 1998). Water temperature was maintained at 20–21°C,and animals were fed a mixture of live zebra fish and frozen shrimptwice a day. Test subjects were held together unless fighting requiredthe insertion of a Plexiglas® wall to separate them. This study wascarried out over the course of a year and a half with filming occurringduring each of the four seasons. Mantle length was monitoredthroughout the 30 day filming period and most animals added 1–2cmto their mantle length during this period.

Filming: regenerative growthTo analyze chromatophore addition during regeneration, a ~3mmsquare piece of tissue from one swimming fin was excised from

each of seven juvenile cuttlefish. This was accomplished by firstremoving animals from their home tanks and placing them in a glassdish with home tank seawater. Animals were anesthetized by slowlyadding 95% ethanol to the seawater. The anesthetic was administeredslowly until the animal’s breathing pattern, color and response totouch indicated that it was sufficiently sedated for filming. A ~3mmsquare of dermis was removed from the anterior fin using anglediris dissection scissors with the aid of stainless steel forceps(Dumont #5 Dumoxel). After surgery, each animal was transferredto a round, glass dish (65�28cm diameter�height) with clean tankwater and returned immediately to their home tank. Including timeto fully anesthetize the animal, the entire surgical procedure took1–1.5h to complete.

Re-growth of the fin dermis into the excised region was monitoredby filming every 3–4days for the duration of the regeneration period,~28–30days, with day 0 being the day the fin tissue was removed.Individuals were removed from their home tank and anesthetizedfollowing the procedure described above for the fin excision surgery.Video images were taken using a flexcam (VideoLabs, Golden Valley,MN, USA) connected to a digital camera (Sony NP-F750, Tokyo,Japan) on a dissecting microscope (Leica Wild M3C, Wetzlar, Hesse,Germany). Because ethanol anesthesia caused the cuttlefish musclesto relax, video recordings and ICD measurements were taken on fullycondensed chromatophores. Lighting and magnification levels wereadjusted during each filming to maximize image clarity and visibilityof chromatophores. Directed light (Fiberlite, Dollan Jenner,Boxborough, MA, USA) was used when needed. Still images wereobtained from video recordings using iMovie (Apple) and WindowsMedia Player (Microsoft) software.

Filming: normal chromatophore growthSeven cuttlefish were filmed three times a week for a period of30days using a charge-coupled device (CCD) digital microscopecamera (Moticam 2000, Richmond, BC, Canada) centrallypositioned within the eyepiece of a dissection microscope (WildM3C). Animals were individually brought from their 473l hometank to the filming area in a small seawater-containing dish andwere lightly anesthetized using 95% ethanol. Anesthesia wasmonitored as described in the previous section. Following sedation,the animal was transferred to a shallow, rounded glass filming dish(41�16cm diameter�height) filled with seawater. The dishcontained an oval groove made within a silicone base (Sylgard 182elastomer) to hold the body of the cuttlefish ventral side down. Thisposition allowed the fin to be raised and flattened across a marked1�1mm grid to standardize measurements during filming.

Data were collected from observations on a 2�2mm section ofthe posterior fin. This location was chosen because it is an area withvery little movement, is easily identifiable, and is far from theanimal’s eyes and head, making filming under lights less stressfulfor the test subject. Using bright directed light from a fiberopticlight source (Fiberlite), video was taken at 20 framess–1 with aMoticam 2000 CCD camera attached to a Wild M3C dissectingmicroscope at a magnification of �400. Three or four adjacentchromatophores were chosen as visual landmarks on the first dayof filming for each animal and each subsequent day of filmingcaptured these selected chromatophores in the same frame forcomparative purposes. This allowed the same 2�2mm area of finto be unequivocally identified during filming. Video from each day’sfilming was transferred to a computer and individual still imagesfrom the video were captured using Windows Media Player and asnipping program. Captured still frames were printed out in colorand used for data analysis.

J. Yacob and others


3425Chromatophore growth and maturation in cuttlefish

Data collection and analysisControl data

For each day of filming, 100 randomly selected blackchromatophores on the same fin as the experimental data but awayfrom the excision/regeneration site were identified and mean inter-chromatophore distance (ICD) computed. ICD was used as ameasure of spatial frequency to quantify the relative distancebetween each chromatophore and its neighbors. Using the stillimages obtained from the filming of normally developing orregenerating fin tissue, mean ICD for each black chromatophorewas determined by measuring the mean distance in millimetersbetween its center and the center of each of its three nearestneighboring black chromatophores. Raw measurements were takenusing Adobe Photoshop CS. These data were then transferred toMicrosoft Excel for data analysis. This same procedure was alsoused to determine the mean daily control ICD for 100 yellow and100 reddish-brown chromatophores using the three closestchromatophores of the same color. Separate control data sets werecollected for the normal growth and regeneration parts of this study.

Regeneration data acquisitionFor each day of filming, mean ICDs in millimeters were obtained forevery new chromatophore when it first appeared in the regeneratingpiece of fin tissue. One-hundred existing chromatophores of the samecolor were randomly selected each filming day from regions adjacentto the excision site and used for control data. Selection of adjacentchromatophores was avoided so that all control data were independent.For each of the new and control chromatophores, the distances to itsthree closest neighboring chromatophores of the same color wererecorded and averaged to find the mean ICD. When comparing stillphotographs from different days, obvious chromatophore clusters andleucophores were used as recognizable markers around the excisedregion. This process made it easier to identify new chromatophores.Because there was slight variance in the ICD of chromatophoresclosest to the edge of the fin compared with more interior

chromatophores, control chromatophores were randomly selectedfrom all regions of the fin surrounding the cut site to avoid skewedcontrol measurements. All data were statistically analyzed using aone-way ANOVA in Microsoft Excel.

Normal maturation data acquisitionEach new black, reddish-brown and yellow chromatophore in the2�2mm filmed region was individually identified and trackedchronologically through to day 30. Every new chromatophore wasgiven a number in the order of its initial appearance in the skin.Nearly all new chromatophores first appeared as a faint orange colornot recognizable as any of the three chromatophore colors (seeResults and Discussion). After several days, before any substantialgrowth in chromatophore size had occurred, each newchromatophore became unmistakably recognizable as black, reddish-brown or yellow. ICD measurements for every new chromatophorewere obtained using chromatophores of that same color class andcollected on the day of its initial appearance only after the color ofa new chromatophore was permanently identifiable. All data werestatistically analyzed using one-way ANOVA in Microsoft Excel.

RESULTSChromatophore insertion during regeneration

Coleoid cephalopods have the very rare ability to completelyregenerate their dermis after substantial damage (Sereni and Young,1932). For example, following excision of a section of the fin froma juvenile S. officinalis, fin re-growth occurs quickly and the newgrowth becomes fully functional and integrated with surroundingtissues within roughly 30days (Fig.1). We used this rapidregeneration ability as a tool to identify a few principles involvedin chromatophore addition. Re-growth and placement ofchromatophores were monitored following the removal of a smallpiece of fin tissue from juvenile S. officinalis as described inMaterials and methods. This experiment focused exclusively on theaddition of black chromatophores. Unequivocal identification of

Fig.1. Time course of regeneration of an excised pieceof fin from a juvenile European cuttlefish, Sepiaofficinalis. A 2–3mm piece of tissue was excised on day0 as described in Materials and methods. After10–12days the muscle and skin tissue layers werealmost completely regenerated. By day 30 it was verydifficult to distinguish the newly re-grown patch of skinfrom surrounding tissue. The excised region was stillnoticeable upon careful inspection because theunderlying iridophores and leucophores had not yet re-grown by day 30. Grid work is laid out in 1mm squares.


3426

individual new black chromatophores was facilitated by the presenceof existing black chromatophores because the latter did not migrateor change location in relation to new chromatophores and becausethey were easily used, along with stationary leucophores, asrecognizable markers around the excised region.

The results of the regeneration experiments demonstrate asignificant difference between the mean ICD of control (i.e. existing)and newly added black chromatophores (Figs2 and 3). Mean ICDsfrom newly inserted chromatophores were substantially greater thanthose of controls during the first few days of this study and remainedso for the first two-thirds of the observation period (days 0–16).Beginning on day 20, mean ICDs from new chromatophoresbecame smaller than control values and this relationship continueduntil the fin was fully regenerated (day 30). Over the 30dayobservation period, the mean ICD of newly added blackchromatophores declined by 44%, starting at an initial value of340m on day 4 and ending with a mean ICD of 190m on day30 (Figs2 and 3). There was very little variation in the mean dailyICDs of existing chromatophores through the course of the study(±20m s.e.m.). A comparison of the mean ICDs from newlyinserted and existing black chromatophores revealed a significantdifference between the two values for each day of this study (P<0.01for all days except day 30 where P<0.05).

Several observations during the regeneration study were noted.Regenerated chromatophores appeared only after tissue re-growth andnot simultaneously. New tissue growth began almost immediately,however, with new tissue appearing after only 24h. Newchromatophores were visibly identifiable for every recorded day,including the first day of data collection. This observation differs fromthat of leucophores, which were not fully restored in the excision sitetissue by the end of the filming period. Newly inserted chromatophoreswere initially observed along the growing edge of the excised region,furthest from existing chromatophores (Fig.1). Only after substantialtissue re-growth had occurred were new chromatophores detected inmore interior regions of the excision site (Fig.1). New chromatophorestended to emerge in groups of approximately 3–5 chromatophores,though single ones did appear in more interior regions of theregenerating piece of fin. It was also observed that the density ofblack chromatophores in the excision site increased duringregeneration until it was first equivalent to and then greater than thatof adjacent, undisturbed fin tissue (Figs2 and 3).

Chromatophore addition during normal growth and maturationTo determine whether the rules for new chromatophore insertionduring regeneration also pertained to normal development, ICDs of

newly added fin chromatophores were collected during normalgrowth and maturation and compared with ICDs from existing(control) chromatophores. Data for each chromatophore type(yellow, reddish-brown and black) were obtained from sevenjuvenile cuttlefish and analyzed separately. No new chromatophorewas used as a data point for ICD analysis until its color becameunequivocally distinguishable as either yellow, reddish-brown orblack.

J. Yacob and others

0 4 6 9 13/14 16 20 23 27 30

450

400

350

300

250

200

150

100

50

0

ControlExperimental

** ** **** ** ** ** ** *

Fig.2. Mean daily inter-chromatophoredistance (ICD) of black chromatophores fromjuvenile European cuttlefish, S. officinalis,during regeneration. Mean ICDs (m) fromnewly added (solid bars) or existing (hatchedbars) chromatophores were obtained everysecond or third day for a 30day datacollection period. Each solid bar representsthe mean ICD from 33–167 newchromatophores and each hatched barrepresents the mean ICD from 400–500existing chromatophores. Error barsrepresent ±1 s.e.m. Asterisks indicate astatistically significant difference between themean daily ICD of newly added and existingchromatophores for that day (*P<0.05 or**P<0.01). NB Data from days 13 and 14were combined.

7

6

5

4

3

2

1

010 60 110 160 210 260 310 360 410 460

Freq

uenc

y

109876543210

181614121086420

10 60 110 160 210 260 310 360 410 460

10 60 110 160 210 260 310 360 410 460

ICD (µm)

A

B

C

Fig.3. Frequency histogram of ICD measurements from newly added blackchromatophores during regeneration in juvenile European cuttlefish, S.officinalis. Data for this figure were taken from Fig.2, days 6, 16 and 30.Black and white arrows indicate the mean ICD from new (experimental)and existing (control) chromatophores, respectively, for each day.



As a general rule, approximately half of the new chromatophoreswere inserted along the outside edge of existing chromatophoresand the remainder were added to the interior between pre-existingchromatophores. The overall mean ICD of newly inserted blackchromatophores was 150±22m (s.e.m.) over the 30daymeasurement period (Fig.4A, Fig.5). In contrast, the overall meanICD for existing (control) black chromatophores was significantlylarger at 190±36m (Fig.4A, Fig.5). Daily mean ICDs of new orexisting chromatophores did not vary significantly during the datacollection period.

The same general pattern was observed for the reddish-brown(Fig.4B, Fig.5) and yellow chromatophores (Fig.4C, Fig.5). Newreddish-brown chromatophores were added at a mean ICD of190±34m compared with their mean control ICD of 210±12m,and new yellow chromatophores appeared at a mean ICD of110±23m compared with their control data of 150±46m(Fig.5). Statistical analysis using a one-way ANOVA showed thatexisting (control) black, yellow and reddish-brown chromatophoreICDs were significantly different from each other (P<0.01). ICDs

of control and new chromatophores were also significantlydifferent from each other for each chromatophore color type(Fig.5), with the ICDs of existing chromatophores being greaterthan those of their respective newly added chromatophores. Theratios of the mean ICDs of new and existing chromatophores foreach color type revealed that new chromatophores were alwaysinserted closer to their nearest neighbors than the normalchromatophore spatial frequency (0.74, 0.85 and 0.57 for black,reddish-brown and yellow chromatophores, respectively). Thesedata also indicate a higher density (i.e. smaller ICD) of yellowchromatophores in the fin than of either black or reddish brownchromatophores (Fig.4C, Fig.5).

Maturation of new chromatophoresTo understand the maturation process of newly formedchromatophores, we individually tracked 516 chromatophores fromtheir initial appearance to their adult phenotype. Newchromatophores always appeared first as very small, pale orangecells, and were not classifiable as any of the three mature

250

200

150

100

50

0

250

200

150

100

50

0

0 2 4 7 9 11 14 16 18 21 23 25 28 30

0 2 4 7 9 11 14 16 18 21 23 25 28 30

** ** * ** ** ** ** ** ** **

0 2 4 7 9 11 14 16 18 21 23 25 28 30

* ** ** ** ** ** ** ** ** ** ** ** **

ICD

(µm

)

Day

A

B

C

Control

Experimental

Control

Experimental

Control

Experimental

250

200

150

100

50

0

** ** ** ** ** ** ** ** ** ** ** ****

Fig.4. Mean daily ICD of black (A), reddish-brown (B) and yellow (C) chromatophores from the European cuttlefish, S. officinalis, during normal growth.Mean ICDs (m) from newly added (solid bars) or existing (hatched bars) chromatophores were obtained every second or third day for a 30day period.Each solid bar represents the mean ICD from 8–30 new chromatophores and each hatched bar represents the mean ICD from 100 existingchromatophores. Error bars represent ±1 s.e.m. Asterisks indicate a statistically significant difference between the mean daily ICD of newly added andexisting chromatophores for that day (*P<0.05 or **P<0.01).


3428

chromatophore colors (Fig.6). Following their initial appearance,all new chromatophores slowly darkened and, in the case of blackchromatophores, grew larger over the course of 27–30days untilthey were the same color and size as existing chromatophores ofthe same color type. For example, black chromatophores turned froman initial pale orange to a reddish-orange color, then to dark brownand finally to their mature dark brown–black color, similar to thatof surrounding black chromatophores (Fig.7). This darkeningdeveloped slowly over the course of a few weeks rather than in astep-wise manner in which color changed dramatically from day today. During the time they were darkening, developing blackchromatophores were never recognizable as either yellow or reddish-brown chromatophores. New reddish-brown and yellowchromatophores developed similarly to black chromatophores,starting out as small pale orange cells that changed color until theyclosely resembled surrounding chromatophores of the same colortype. Once a chromatophore became recognizable as a fully grownblack, reddish-brown or yellow chromatophore, its color remainedunaltered for the remainder of the observation period.

To quantify the changes in size as chromatophores matured, thediameters of new chromatophores were measured over time as theydeveloped. The diameters of existing yellow, reddish-brown andblack chromatophores were also obtained as controls. Because theethanol anesthesia procedure causes relaxation of the peripheralmusculature including the chromatophore muscles, allchromatophore diameter measurements were performed oncompletely condensed chromatophores. Newly inserted black

chromatophores grew substantially as they developed and matured,enlarging from an initial mean diameter of 17±0.39m on day 1 toa diameter of 36±0.57m on day 29 (Fig.8A). Unlike their blackcounterparts, new reddish-brown and yellow chromatophores didnot significantly change in size during maturation, with theirdiameters remaining statistically similar to those of existingchromatophores throughout the study (diameters of 18±0.32 and18±0.44m, respectively; Fig.8B,C). On the first day of appearance,the diameters of new reddish-brown, yellow and blackchromatophores were all statistically similar to each other (P>0.05).

DISCUSSIONNew chromatophore insertion

The objective of this study was to identify a few of the principlesunderlying chromatophore organization during normal growth. Thisis a necessary first step towards understanding how the centralnervous system mediates body patterning and how chromatophoresare organized and arranged in the skin of cephalopods for the twinpurposes of camouflage and communication. Chromatophores,when appropriately distributed across the body, are the primarycomponents of the numerous, intricate, body patterns that mediateintraspecies and interspecies communication, including camouflage,mating and prey capture (Hanlon and Messenger, 1988; Boal et al.,2004). In contrast, an irregular or patchy arrangement ofchromatophore cells during growth would interfere with the properproduction of body patterns, and negatively impact the survival andreproductive success of these organisms. The results of this study

J. Yacob and others

250

200

150

100

50

0ExistingN=9800

NewN=225

ExistingN=2310

NewN=108

ExistingN=7700

NewN=183

190.3

152.0

213.6193.0

146.0

107.0

ICD

(µm

)

**†† **

††

**††

Fig.5. Summary of mean ICD of existing and new black (left),reddish-brown (middle) and yellow (right) chromatophores fromjuvenile European cuttlefish, S. officinalis, during 30days ofnormal growth. Solid bars indicate overall mean ICDs for newchromatophores and hatched bars represent data from existingchromatophores. Error bars represent ±1 s.e.m. Asterisksindicate a statistically significant difference between the meanICD of new and existing chromatophores (**P<0.01). Statisticalanalysis using a one-way ANOVA found that existing (control)black, yellow and reddish-brown ICDs were significantly differentfrom each other (††P<0.01).

Fig.6. Time course of chromatophore appearance andmaturation in the fin of a juvenile European cuttlefish,S. officinalis, during normal growth. This figure showsthe addition of seven new chromatophores to theposterior fin of a juvenile S. officinalis during the 30dayobservation period. Letters A–G show the location ofnew chromatophores and red circles indicate the day oftheir first appearance. Labels 1–11 show the location ofexisting chromatophores. Two existing chromatophores,nos 5 and 8, initially had the appearance of orangechromatophores and over time became darker untilthey were the same size and color as surroundingblack chromatophores. Another chromatophore, no. 11,was observed part-way through this maturationprocess. The time course in days refers to the 30daydata collection period as described in Materials andmethods.



strongly support the hypothesis that the chromatic component ofbody patterning in the European cuttlefish, S. officinalis, is highlyorganized and that the organizational aspects of chromatophoregrowth (i.e. density, positioning) are precisely regulated.

Two observations from the fin regeneration study suggest thatchromatophore addition may be more actively controlled duringdermal regeneration than other dermal elements such as leucophoresand iridiphores. First, new chromatophores in regenerating tissueappeared shortly after general tissue growth occurred, in less thana week (Figs1–3). In contrast, new leucophores and iridophores didnot appear until the end of the 30day study and were not fullyregenerated once all chromatophores had re-grown and the studyhad been completed. This observation suggests the mechanism(s)responsible for chromatophore growth and subsequent innervationduring regeneration is either continuously active or activated inimmediate response to tissue damage.

The second indication of active regulation was the progressivedecrease in the mean experimental ICD during tissue regenerationuntil it was equal to and then less than that of the surrounding tissue(Figs2 and 3). The first sets of new chromatophores were insertedin the interior of the regeneration site, relatively far from theirneighbors. Subsequent new chromatophores filled in the gapsbetween the first set of chromatophores until the ICD of newchromatophores in the regeneration site was less than that in controltissue. This observation was similar to that of the normal growthstudy in which new chromatophores were added at ICDs less than

those in controls. This suggests that at this point in regeneration thetissue was acting similarly to normally growing tissue. The acutereduction in the ICD of new chromatophores during regenerationsuggests that more than one mechanism may be involved indetermining the location of a new chromatophore. If only oneprimary factor was responsible for new chromatophore positioning,the ICDs of new chromatophores would be the same as control ICDsthroughout the entire regeneration process. In such a model, newchromatophores would be added linearly only along the growingedge of the regeneration site close to existing chromatophores.Instead, by adding chromatophores to the interior of the regenerationsite at large ICDs and then filling in the gaps, the regenerating sitebecomes functional sooner than would have been possible in thesingle mechanism model. To facilitate this more efficientregeneration strategy, it is suspected that the positioning of newchromatophores is controlled by both positive and negativeregulators. The presence of positive regulators promoteschromatophore growth and innervation, while negative regulatorsinhibit chromatophores from developing too close to otherchromatophores, thus maintaining a relatively consistent averageICD. The predictable manner in which new chromatophores areinserted relative to existing chromatophores indicates a high levelof organization and strong positive evidence that the presence ofold, neighboring chromatophores helps determine the location ofnew chromatophores. This top-down model is consistent with theevidence presented here; however, it must be noted that othermodels, including a simple attraction–repulsion model, are alsopossible.

In normally growing tissue, most new chromatophores wereadded in between existing chromatophores. Some newchromatophores, however, were also inserted along the outside edgeof the fin to complete lines and fill in gaps as the fin enlarged. Newchromatophores inserted in this manner rarely appeared far awayfrom existing chromatophores and had ICDs close to or less thancontrols (J.Y., personal observation). Consistent with the results ofthe regeneration study, this observation provides evidence that theproximity of existing chromatophores is one of the primary effectorsregulating the location of new chromatophores. The addition of newchromatophores at smaller than normal ICDs presumably providesfor growth without reducing the grain density of chromatophoresin body patterns.

Development of new chromatophoresThe details of chromatophore development in cephalopods remainunresolved, in large part because of a paucity of studies. Twocompeting hypotheses have been proposed. One study in Octopusvulgaris proposed that all new chromatophores begin as yellowchromatophores, and then turn reddish-brown before finallybecoming black chromatophores (Packard, 1985). An alternativehypothesis is that each chromatophore is born as a specific colortype and remains that type, e.g. yellow, reddish-brown or black, forits entire lifetime. The data presented in this paper refute the firsthypothesis and provide support for the second (Fig.9).

The majority of new chromatophores observed in this studystarted out as small, pale orange cells that developed directly intoyellow, reddish-brown or black chromatophores without passingthrough either of the other two colors during the month-longobservation period. We looked very closely but did not observe anyreddish-brown chromatophores passing through a transient yellowphase or a mature black chromatophore which began yellow in colorand went through a reddish-brown phase before becoming black.All yellow chromatophores observed on the first day of filming

Fig.7. Ontogeny of black chromatophores during normal growth in theEuropean cuttlefish, S. officinalis. The figure shows the initial appearanceand subsequent maturation of three black chromatophores. The left panelillustrates one chromatophore from its first day of appearance (day 2) untilthe end of the data collection period (day 30). The right panel shows theappearance and maturation of two black chromatophores (nos 1 and 3).Each new chromatophore starts as a pale orange color and slowly darkensand enlarges over the course of a few weeks until it is the same size andcolor as neighboring black chromatophores. At no stage does thechromatophore appear as a fully grown yellow or reddish-brownchromatophore. Note the presence in the right panel of a fully mature blackchromatophore (no. 2) throughout the data collection period and of two neworange-colored chromatophores that appear on days 25 and 30,respectively. The time course in days refers to the 30day data collectionperiod as described in Materials and methods.


3430

remained yellow at the end of the 30day observation period, duringwhich we observed the appearance of many new blackchromatophores. The observation that yellow chromatophores didnot darken while many new orange-colored chromatophores becamefully mature black chromatophores is strong evidence against theidea that groups of chromatophores develop together in waves andall mature together at the same rate while passing through a yellow,

reddish-brown and black developmental sequence as proposed byPackard (Packard, 1985). Our data support Packard’s observationthat new chromatophores are always small and orange-colored(Packard, 1985).

An important observation was that those new chromatophoresfated to become black chromatophores doubled their diameters overthe course of a few weeks (Fig.8) while the yellow and reddish-

J. Yacob and others

** ** ** ** ** ** ** ** ** ** **D

iam

eter

(µm

)

Day

A

B

C

Control

Experimental

Control

Experimental

Control

Experimental

1 3 6 8 10 12 15 17 20 22 24 27 29

45

40

35

30

25

20

15

10

5

0

1 3 6 8 10 12 15 17 20 22 24 27 29

1 3 6 8 10 12 15 17 20 22 24 27 29

30

25

20

15

10

5

0

30

25

20

15

10

5

0

Fig.8. The diameters of existing (hatched) and new (solid) chromatophores during maturation in the European cuttlefish, S. officinalis. (A)It was observedthat black chromatophores grew substantially over time, changing from a mean (±s.e.m.) diameter of 17±0.39m when they first appeared to 36±0.57m onday 29. (B)Unlike black chromatophores, reddish-brown chromatophores did not greatly increase in size and remained very similar in diameter to controlchromatophores throughout the data collection period. The mean diameter of reddish-brown control and experimental chromatophores was 18±0.32m.(C)Similar to reddish-brown chromatophores, yellow chromatophores did not greatly increase in size and remained very similar in diameter to controlchromatophores throughout the study. The mean diameter of yellow control and experimental chromatophores was 18±0.44m.



brown chromatophores remained the same size as at their initialappearance. The larger size of the black chromatophores may bedue to the need to generate large areas of black on the skin for stripedpatterns such as the zebra and passing cloud displays (Packard andSanders, 1971; Hanlon and Messenger, 1996; Hanlon andMessenger, 1988). Each chromatophore color type lies in its owndermal layer, with the black chromatophores being the mostsuperficial followed by the reddish-brown and yellowchromatophore layers (Cloney and Florey, 1968; Cloney andBrocco, 1983; Hanlon and Messenger, 1996; Sutherland et al., 2008).This study was unable to track any vertical movement ofchromatophores as they matured. However, given the knowledgethat chromatophores of different colors reside in separate dermallayers and the hypothesis proposed by Packard (Packard, 1985) thatchromatophores pass through all three colors, this would be aninteresting topic for future investigations.

Principles of chromatophore insertion and maturation in thecuttlefish S. officinalis

Our data show that chromatophore growth and development arehighly organized and structured. From our observations, we proposefive guiding principles of chromatophore growth and maturation.(1) Each of the three chromatophore cell types – black, reddishbrown and yellow – are present at different spatial frequencies inthe cuttlefish fin. (2) During normal growth, new chromatophoresare inserted at a higher spatial frequency than existing (control)

chromatophores of the same color type. (3) In regenerating tissue,new black chromatophores are initially added at low spatialfrequencies. As regeneration continues, new black chromatophoresappear at increasing spatial frequencies until they are inserted at aspatial frequency higher than observed in control tissue. (4) Allchromatophores first appear as pale orange cells and slowly darkeninto their respective color types without passing through intermediatecolor stages. (5) New black chromatophores undergo a doubling insize as they mature while, reddish-brown and yellowchromatophores do not grow at all after they are inserted in thedermis.

Several of these principles have been observed in other studies(Packard, 1985; Hanlon and Messenger, 1988; Messenger, 2001;Domingues, 2001); however, this is the first study to statisticallyanalyze the growth and spatial organization of chromatophores asthey develope. The use of modern data capture and analysestechniques allowed a vastly larger number of chromatophores to betracked and analyzed than was possible in earlier studies. A well-developed understanding of the rules controlling chromatophoreaddition and positioning is an important first step in elucidating thebasic neural organization and processes that underlie the amazinglyhigh level of behavioral plasticity in cephalopods. Although thisstudy did not analyze new chromatophore motoneurons or theirinnervation pattern, the principles in this study will be a helpfulstarting point for future examination of this issue. By using theprinciples outlined in this study it should be possible to makeaccurate statistical predictions of the location of new chromatophoreson the fin. This information should make it easier to predict thelocation of new neurons and innervations and observe and analyzethem as they grow.

Further elucidation of the underlying molecular mechanismscontrolling chromatophore insertion and spacing is required to obtaina full understanding of how chromatophore growth takes place andhow new chromatophores are integrated within the central nervoussystem. Many questions still remain about how chromatophoresdevelop and mature. Do all three color cells begin as a pluripotentcell that responds to external factors to differentiate into one of thethree colors, or are there three separate cell types that lead to eachcolor? Do new chromatophore cells divide close to the site at whichthey are needed, or do they migrate from other parts of the body?Because many cephalopod species have varying densities ofchromatophores (Cloney and Florey, 1968; Packard and Hochberg,1977; Cloney and Brocco, 1983; Hanlon and Messenger, 1996;Sutherland et al., 2008), it would also be interesting to see whetherthe patterns and principles observed in S. officinalis in this studyare conserved across other cephalopod species.

ACKNOWLEDGEMENTSWe thank Dr Zhuobin Zhang and Mr Eli Goodwin for their thoughtful comments onearly versions of this manuscript. This material is based upon work supported bythe Air Force Office of Scientific Research under award no. FA9550-09-1-0395.

REFERENCESBarbosa, A., Mäthger, L. M., Chubb, C., Florio, C., Chiao, C. C. and Hanlon, R. T.

(2007). Disruptive coloration in cuttlefish: a visual perception mechanism thatregulates ontogenetic adjustment of skin patterning. J. Exp. Biol. 210, 1139-1147.

Boal, J. G., Shashar, N., Grable, M. M., Vaughan, K. H., Loew, E. R. and Hanlon,R. T. (2004). Behavioral evidence for intraspecific signaling with achromatic andpolarized light by cuttlefish. Behaviour 141, 837-861.

Boycott, B. B. (1961). The functional organization of the brain of cuttlefish Sepiaofficinalis. Proc. R. Soc. Lond. B 153, 503-534.

Brocco, S. L. and Cloney, R. A. (1980). Reflector cells in the skin of Octopus dofleini.Cell Tissue Res. 205, 167-186.

Carvalho, L. N., Zuanon, J. and Sazima, I. (2006). The almost invisible league:crypsis and association between minute fishes and shrimps as a possible defenceagainst visually hunting predators. Neotrop. Ichthyol. 4, 219-224.

Mature chromatophores

Yellow

New chromatophores

Black

Reddish-brown

Fig.9. Schematic model of chromatophore maturation in the Europeancuttlefish, S. officinalis. All new chromatophores first appear as small, paleorange cells not recognizable as any of the three color types. Over thecourse of a few weeks black chromatophores gradually darken, starting outas small orange-colored cells and passing through a reddish-orange phase,then dark brown, and finally to their mature dark brown–black color. Theyalso increase in size until they closely resemble neighboring blackchromatophores. Yellow and reddish-orange chromatophores darken totheir respective colors but do not grow significantly in size as they develop.Black chromatophores are never observed to pass through the yellow orreddish-orange color types during maturation.


3432

Cloney, R. A. and Brocco, S. L. (1983). Chromatophore organs, reflector cells,iridocytes and leucophores in cephalopods. Am. Zool. 23, 581-592.

Cloney, R. A. and Florey, E. (1968). Ultrastructure of cephalopod chromatophoreorgans. Zeitschr. Zellforsch. 89, 250-280.

Cott, H. B. (1940). Adaptive Coloration in Animals. London: Metheun & Co.Domingues, P. M. (2001). Growth of young cuttlefish, Sepia offcinalis (Linnaeus 1758)

at the upper end of the biological distribution temperature range. Aquacult. Res. 32,923-930.

Fox, H. M. and Vevers, G. (1960). The Nature of Animal Colours. New York:Macmillan Press.

Hanley, J. S., Shashar, N., Smolowitz, R., Bullis, R. A., Mebane, W. N., Gabr, H. R.and Hanlon, R. T. (1998). Modified laboratory culture techniques for the Europeancuttlefish Sepia officinalis. Biol. Bull. 195, 2, 223-225.

Hanlon, R. T. (2007). Cephalopod dynamic camouflage. Curr. Biol. 17, 400-404.Hanlon, R. T. and Messenger, J. B. (1988). Adaptive coloration in young cuttlefish

(Sepia officinalis): the morphology and development of body patterns and theirrelation to behavior. Philos. Trans. R. Soc. Lond. 320, 437-487.

Hanlon, R. T. and Messenger, J. B. (1996). Cephalopod Behaviour. Cambridge:Cambridge University Press.

Hebert, P. (1974). Spittlebug morph mimics avian excrement. Nature 150, 352-354.Johnsen, S. (2001). Hidden in plain sight: the ecology and physiology of organismal

transparency. Biol. Bull. 201, 301-318.Loi, P. K. and Tublitz, N. J. (1998). Long term rearing of cuttlefish in a small scale

facility. Aqua. Sci. Conserv. 2, 1-9.

Messenger, J. B. (2001). Cephalopod chromatophores: neurobiology and naturalhistory. Biol. Rev. 76, 473-528.

Packard, A. (1985). Sizes and distribution of chromatophores during post-embryonicdevelopment in cephalopods. Vie Milieu 35, 285-298.

Packard, A. and Hochberg, F. G. (1977). Skin patterning in Octopus and othergenera. Symp. Zool. Soc. Lond. 38, 191-231.

Packard, A. and Sanders, G. (1969). What the octopus shows to the world.Endeavour 28, 92-99.

Packard, A. and Sanders, G. (1971). Body patterns of Octopus vulgaris andmaturation response to disturbance. Anim Behav. 19, 780-790.

Sazima, I., Carvalho, L. N., Mendonça, F. P. and Zuanon, J. (2006). Fallen leaveson the water-bed: diurnal camouflage of three night active fish species in anAmazonian streamlet. Neotrop. Ichthyol. 4, 119-122.

Sereni, E. and Young, J. Z. (1932). Nervous degeneration and regeneration incephalopods. Pubbl. Staz. Zool. Napoli 12, 173-208.

Severaid, J. (1945). Pelage changes in the snowshoe hare (Lepus americanusstruthopus Bangs). J. Mammal. 26, 41-63.

Shohet, A., Baddeley, R., Anderson, J. and Osorio, D. (2007). Cuttlefishcamouflage: a quantitative study of patterning. Biol. J. Linn. Soc. 92, 335-345.

Stevens, M. and Merilaita, S. (2009). Animal camouflage: current issues and newperspectives. Philos. Trans. R. Soc. B 364, 423-427.

Sutherland, R. L., Mathger, L. M., Hanlon, R. T., Urbas, A. M. and Stone, M. O.(2008). Cephalopod coloration model. I. Squid chromatophores and iridophores. J.Opt. Soc. Am. A 25, 2044-2054.

J. Yacob and others


RESEARCH ARTICLE Principles underlying chromatophore ... · Accepted 1 June 2011 SUMMARY The goal...

Documents

Transcript of RESEARCH ARTICLE Principles underlying chromatophore ... · Accepted 1 June 2011 SUMMARY The goal...