Relationship between body colour, feeding,and reproductive arrest under short-day developmentin Tetranychus pueraricola (Acari: Tetranychidae)

Katsura Ito • Tatsuya Fukuda • Hiroshi Hayakawa • Ryo Arakawa •

Yutaka Saito

Received: 16 May 2012 / Accepted: 30 January 2013 / Published online: 19 February 2013� Springer Science+Business Media Dordrecht 2013

Abstract In Tetranychus spider mites (Acari: Tetranychidae), diapausing females have a

conspicuous orange body colour, which is used as an indicator of diapause induction in

many laboratory studies. However, to which extent body colour reflects reproductive

activity is scarcely investigated. In this study, we investigated the relationship between

body colour, reproductive arrest, and food intake in the inbred strain of T. pueraricola

individually reared at 20 �C with a 10:14 h light: dark photoperiod. Our results showed

that (1) body colour is a good indicator of reproductive arrest 11 days after adult emer-

gence but does not completely reflect reproductive status at an earlier age; (2) even orange

females intermittently feed, and the arrest of feeding comes after the change in body

colour; and (3) reproducing females have a higher risk of death than non-reproducing

females. These results suggest that measurement of diapause incidence by body colour

alone may miss the variation in reproductive status in early adult life.

Keywords Acari � Diapause incidence � Environmental variation � Spider mite �Tetranychidae

Introduction

In Tetranychus spider mites, the body colour of the females developed under short-day and

low-temperature conditions becomes conspicuous orange. This is a phenotype involved in

the adult diapause of spider mites (reviewed in Veerman 1985), and thus has been used as

an indicator of diapause induction in many experimental studies (e.g. Veerman 1977;

K. Ito (&) � T. Fukuda � R. ArakawaFaculty of Agriculture, Kochi University, Monobeotsu 200, Nankoku, Kochi, Japane-mail: ktr@kochi-u.ac.jp

H. HayakawaNational Institute for Agro-Environmental Science, 3-1-3 Kanondai, Tsukuba, Japan

Y. SaitoGraduate school of Agriculture, Hokkaido University, 9 nishi 9 kita, kita-ku, Sapporo, Japan

123

Exp Appl Acarol (2013) 60:471–477DOI 10.1007/s10493-013-9660-3

Ignatowicz and Helle 1986; Goka and Takafuji 1990, 1991; Takafuji et al. 1991; Kroon

et al. 1998; Koveos et al. 1999; Ito 2003; Kawakami et al. 2009; Suzuki and Takeda 2009).

However, what ‘diapause’ means in them is often unclear, because the change of colour

is only one of the physiological attributes in diapause. The most important character in the

ecological context is reproductive arrest. Nevertheless, to which extent the colour change

reflects the reproductive activity has scarcely been investigated. As Saito et al. (2005)

pointed out, the most reliable criterion to determine reproductive diapause is the pattern of

reproduction itself, and the body colour is useful only after this is verified to be parallel to

the cessation of reproduction. Recent studies suggest that the arrest of egg deposition is not

accompanied by a change in pigmentation in Stigmaeopsis species (Saito et al. 2002,

2005). Following field observation showed that the change in pigmentation occurs

[1 month after the cessation of egg production in S. longus in southwestern Japan (Ito

unpublished data). These results suggest that body pigmentation and reproductive arrest

might generally be governed by separate physiological mechanisms. Thus, the use of body

colour to infer reproductive status needs to be re-examined.

The aim of this study is to clarify the relationship between body colour, reproductive

arrest, and feeding under constant day length and temperature conditions by using the spider

mite Tetranychus pueraricola Ehara and Gotoh as a model species. This species is considered

to be closely related to the red form of T. urticae (Suwa and Gotoh 2006); moreover, it is

widely distributed throughout Japan and infest kudzu vine (Pueraria lobata [Willd.]) in the

wild, on which it makes conspicuous red scars on the leaf surface due to mass feeding (Gotoh

and Tokioka 1996; Gotoh et al. 2004). This species shows geographic variation in the ten-

dency to ‘diapause’, and crossing experiments with simple Mendelian inheritance have

shown that ‘non-diapause’ is a dominant character over ‘diapause’ (Suwa and Gotoh 2006).

However, these conclusions are based on the body colour, so that the correlation between

body colour and reproduction is unknown. To enable the clear assessment of environmental

effects, we used a genetically homogeneous inbred line established from 1 female.

Materials and methods

Mites

The strain of T. pueraricola mite used in this study was derived from one female obtained

from a kudzu-infesting population, which was collected from Hitsuzan Park, Hitsuzan,

Kochi, Japan (N 33.551, E 133.536, WGS84) on 7 December 2010. The species was

identified primarily from the characteristic shape of the knob on the aedeagus in males, the

body colour of adults, and the red scars along leaf veins of the host plant (Ehara and Gotoh

2009). The isofemale strain was established by repeated full-sib (brother-sister) mating for

[12 generations on 2 9 2-cm leaf squares of kidney bean (Phaseolus vulgaris L.)

maintained at 25 �C with a 16:8 h light: dark (L:D) photoperiod. Thereafter, individuals

were allowed to mate freely and were kept at 20 �C with a 16:8 h L:D photoperiod. All

rearing and experimental procedures used the ordinary leaf-disk method with water-soaked

cotton pads (Ehara and Gotoh 2009).

Diapause induction

Twenty adult females were placed on a bean leaf and allowed to oviposit for 1 day while

being maintained at 25 �C with a 16:8 h L:D photoperiod. After the removal of females,

472 Exp Appl Acarol (2013) 60:471–477

123

the culture was kept at 20 �C with a 10:14 h L:D photoperiod, which reflected the October

(autumn) field conditions at the collection site of this population.

On days 13 and 14 each, 54 newly emerged adult females were individually isolated on

a 1 9 1-cm leaf square. The survival rate and the day of first oviposition were checked

every day until day 28. Every 7 days, mites were transferred to new leaf squares, and the

number of deposited eggs was recorded. If mites were in reproductive arrest, the presence

or absence of leaf scars (made by feeding) was recorded (on days 14, 21, and 28). Body

colour was checked every day, and the day at which mites became orange was recorded.

Statistical analysis

Survival curves were plotted for all adult females and each colour morph by using the

Kaplan–Meier method. A log-rank test was used to compare the curves of each morph. A

two-way ANOVA was used to test the effects of body colour and week on the egg

production, and Ryan’s method for multiple comparisons of proportions (Ryan 1960) was

used to compare the proportion of feeding females between different weeks. All statistical

analyses were conducted using R version 2.13.0 (R Development Core Team 2011).

Results

30 out of 108 females underwent a change in body colour to orange at some point during

the 28-days observation period, whereas the others remained non-pigmented within the

observation period. The change in body colour began after 7 days of adult life (days

20–21) and become conspicuous until 11 days (days 25–26). The observed orange colour

was distinct from the red colour that starved females exhibit (this is typically much weaker

than the colour of diapausing).

The survival curves of all females and of each colour morph (‘orange’ and ‘non-

orange’) 11 days after adult emergence are shown in Fig. 1. Orange females had a sig-

nificantly lower risk of death than non-orange females (log-rank test, P \ 0.001). By day

11, 33 non-orange females had died, 29 of which produced C1 egg(s) in 2.103 ± 0.143

(mean ± SE) days after adult emergence.

Figure 2 shows the histogram of adult age of first reproduction. Distribution of the day

was bimodal. One group, which mostly included non-orange females, was distributed

around day 2. The other group, which included only orange females, did not oviposit

within 28 days (rightmost bar in Fig. 2). There were 3 intermediate individuals that started

oviposition on days 2–5 but later arrested oviposition and became orange until day 28

(Table 2, oviposition).

The weekly mean number of eggs produced by each colour morph (including 0 counts)

is presented in Fig. 3. The summary of ANOVA is presented in Table 1, showing a

significant effect of only body colour. Non-orange females constantly produced an average

of ca. 25 eggs, whereas orange females produced \5 eggs; A slight increase in egg

production after day 15 was attributed to 2 females that began oviposition within 15 days

(Fig. 2, day 7 and 13 of orange females). Data excluding these females is summarised in

Table 2 (oviposition).

All 23 orange females for which a complete set of feeding data are available were feeding

at some time during the observation period, and of these 4 females fed every week. The

proportion of females feeding significantly decreased during 15–21 days when compared

with the previous week (Table 2, feeding). All non-orange females (n = 78) were feeding.

Exp Appl Acarol (2013) 60:471–477 473

123

Discussion

Orange pigmentation in T. pueraricola became conspicuous 11 days after adult emergence

(days 25–26) at 20 �C with a 10:14 h L:D photoperiod. This period is similar to that in

T. kanzawai, which develops colouration 26 days after emergence (Ito 2003). The females

could be divided into 2 groups depending on the day of first oviposition: in 1 group,

females oviposited on approximately 2 days and were not coloured, while in the other

group, females did not oviposit within 28 days and had orange pigmentation (Fig. 2).

These 2 groups may correspond to the ‘non-diapause’ and ‘diapause’ groups in traditional

experiments.

0 5 10 15 20 25

0.0

0.2

0.4

0.6

0.8

1.0

Days after adult emergence

Sur

viva

l rat

e

10 15 20 25

0.0

0.2

0.4

0.6

0.8

1.0

Days after adult emergence

Sur

viva

l rat

e

Non-orangeOrange

A

B

Fig. 1 A Survival curve of allfemales (solid line) and its 95 %CI (dashed line). B Survivalcurves of 2 colour forms after11 days of adult emergence.Survival rate at day 10 was resetto 100 %. P \ 0.001 (log-ranktest)

1 3 5 7 9 11 13 15 17 19 21 23 25 27

Days after adult emergence

No.

of f

emal

es

0

10

20

30

40

50Non-orangeOrange

*

Fig. 2 Stack histogram showingthe day of first oviposition afteradult emergence. Black barindicates the females thatchanged colour within 28 daysfrom adult emergence(reproduction before colourchange occurred at day 2, 4, and5; see Table 2, oviposition). Thebar on the far right (with asterisk)corresponds to females that didnot oviposit during theobservation period

474 Exp Appl Acarol (2013) 60:471–477

123

However, 3 females laid several eggs before body colouration. Given that diapause

typically cannot be terminated until long after the disappearance of the unfavourable

conditions that induced it (Veerman 1985), these females might not be in diapause in a

Fig. 3 Weekly egg productionby each colour form (including 0counts). Data are presented asmean and SD. Data of 2 females,which started reproducing within2 weeks (days 7 and 13), areincluded in oviposition data.ANOVA table is summarised inTable 1

Table 1 Results of the two-way ANOVA testing the effect of colour form and day on egg production (seeFig. 3)

Source Df SS MS F P

Colour 1 37,607 37,607 218.521 \0.001

Day 3 235 78 0.455 0.714

Interaction 3 151 50 0.292 0.831

Residuals 281 48,359 172

Table 2 Status of feeding and oviposition in orange (diapausing) females during each week

Days

0–7 8–14 15–21 21–27

Feeding

No. of females* – 25 26 24

No. of feeding females – 21 10 15

Prop. feeding** – 0.84A 0.38B 0.63AB

Oviposition

No. of females 28 28 26 26

No. of ovipositing females 3 1 0 0

Prop. ovipositing 0.11 0.04 0.00 0.00

Mean 1.67 1.00 – –

SD 1.16 – – –

Data for 2 females that started to oviposit after 2 weeks (days 7 and 13) are excluded

* Several data are not available

** Columns with different letters were significantly different (P \ 0.05, Ryan’s method)

Exp Appl Acarol (2013) 60:471–477 475

123

strict sense (or quite shallow diapause). In any case, this result indicates that body colour

reflects the reproductive status in the later adult life but does not necessarily reflect that in

the early life. In the congener T. kanzawai, reproductive arrest and a change in body

colouration occurred if day length conditions were changed from long to short at the time

of adult emergence (Saito et al. 2005). The present study indicates that this change can

occur even under constant day length conditions in T. pueraricola. In addition, because

most of the females dying before pigmentation were producing eggs, exclusion of dead

females from experimental count may underestimate the degree of real reproductive

diapause.

Notably, even orange females frequently feed after the development of colouration.

Orange females isolated in cages and preserved in cool conditions tend to desiccate in

T. telarius (McEnroe 1961) and often have lower fecundity following diapause in

T. kanzawai (Ito 2004). Thus, feeding activity might compensate for the loss of body water,

allowing physiological conditions for reproduction to be maintained until diapause has

ended. The relationship between body colour, reproductive arrest, and feeding, as observed

in this study, is shown in Fig. 4.

In mass-rearing experiments, females are allowed to lay eggs for several days and are

maintained under short-day conditions until 1–2 weeks after adult emergence, following

which the proportion of orange coloured females is assumed to reflect the incidence of

diapause. Although such an assumption may be appropriate when evaluating the incidence of

reproductive arrest, it could ignore the variation in reproductive status during early adult life.

Acknowledgments We appreciate Dr. T. Gotoh for kindly providing information on T. pueraricola thisspecies. We thank Dr. H. Numata and Dr. S.G. Goto for providing useful information on insect diapause. Wethank Mr. Y. Che, Mr. Y. Paku, and Ms. K. Tamura for assistance with experiments.

References

Development Core Team R (2011) R: a language and environment for statistical computing. R Foundationfor Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/

Ehara S, Gotoh T (2009) Colored guide to the plant mites of Japan. Zenkoku Noson Kyoiku Kyokai, Tokyo(in Japanese)

Goka K, Takafuji A (1990) Genetical studies on the diapause of the two-spotted spider mite, Tetranychusurticae Koch (1). Appl Entomol Zool 25:119–125

Goka K, Takafuji A (1991) Genetical studies on the diapause of the two-spotted spider mite Tetranychusurticae Koch (2). Appl Entomol Zool 26:77–84

Gotoh T, Tokioka T (1996) Genetic compatibility among diapausing red, non-diapausing red and diapausinggreen forms of the two-spotted spider mite, Tetranychus urticae Koch (Acari: Tetranychidae). Jpn JEntomol 64:215–225

Pigmentation

Reproductivearrest

Cessation offeeding

Fig. 4 Venn diagram aboutrelationship between diapausecharacteristics

476 Exp Appl Acarol (2013) 60:471–477

123

Gotoh T, Suwa A, Kitashima Y, Rezk HA (2004) Developmental and reproductive performance of Tetr-anychus pueraricola Ehara and Gotoh (Acari: Tetranychidae) at four constant temperatures. ApplEntomol Zool 39:675–682

Ignatowicz S, Helle W (1986) Genetics of diapause suppression in the two-spotted spider mite, Tetranychusurticae Koch. Exp Appl Acarol 2:161–172

Ito K (2003) Effect of leaf condition on diapause induction of a Kanzawa spider mite Tetranychus kanzawaiKishida (Acari: Tetranychidae) population on tea plants. Appl Entomol Zool 38:559–563

Ito K (2004) Deteriorating effects of diapause duration on postdiapause life history traits in the Kanzawaspider mite. Physiol Entomol 29:453–457

Kawakami Y, Numata H, Ito K, Goto SG (2009) Dominant and recessive inheritance patterns of diapause inthe two-spotted spider mite Tetranychus urticae. J Hered 101:20–25

Koveos DS, Veerman A, Broufas GD, Exarhou A (1999) Altitudinal and latitudinal variation in diapausecharacteristics in the spider mite Tetranychus urticae Koch. Entomol Sci 2:607–613

Kroon A, Veenendaal RL, Veerman A (1998) Response to photoperiod during diapause development in thespider mite Tetranychus urticae. J Insect Physiol 44:271–277

McEnroe WD (1961) The control of water loss by the two-spotted spider mite (Tetranychus telarius). AnnEntomol Soc Am 54:883–887

Ryan TA (1960) Significance tests for multiple comparison of proportions, variances and other statistics.Psychol Bull 57:318–328

Saito Y, Sakagami T, Sahara K (2002) Differences in diapause attributes between two clinal forms dis-tinguished by male-to-male aggression in a subsocial spider mite, Schizotetranychus miscanthi Saito.Ecol Res 17:645–653

Saito Y, Ito K, Sakagami T (2005) Imaginal induction of diapause in several adult-diapausing spider mites.Physiol Entomol 30:96–101

Suwa A, Gotoh T (2006) Geographic variation in diapause induction and mode of diapause inheritance inTetranychus pueraricola. J Appl Entomol 130:329–335

Suzuki T, Takeda M (2009) Diapause-inducing signals prolong nymphal development in the two-spottedspider mite Tetranychus urticae. Physiol Entomol 34:278–283

Takafuji A, So PM, Tsuno N (1991) Inter- and intra-population variations in diapause attribute of the two-spotted spider mite, Tetranychus urticae Koch, in Japan. Res Popul Ecol 33:331–344

Veerman A (1977) Aspects of the induction of diapause in a laboratory strain of the mite Tetranychusurticae. J Insect Physiol 23:703–711

Veerman A (1985) Diapause. In: Helle W, Sabelis MW (eds). Spider mites. Their biology, natural enemiesand control, 1A. Elsevier, Amsterdam, pp 279–316

Exp Appl Acarol (2013) 60:471–477 477

123