Genetica 97: 197-203, 1996. 197 (~) 1996 Kluwer Academic Publishers. Printed in the Netherlands.

The effect o f B c h r o m o s o m e s on m a t i n g success of the gras shopper Eyprepocnemis plorans

Silvia Martin 1, Pilar Arana 1 & Nuno Henriques-Gil 2. I Departamento de Gendtica, Facultad de Biologfa, Universidad Complutense, Madrid 28040, Spain; 2 Unidad de Gengtica, Universidad San Pablo C.E.U., Monteprfncipe, Boadilla del Monte, Madrid 28668, Spain *Address for correspondence

Received 24 April 1995 Accepted 6 July 1995

Key words: B chromosomes, Eyprepocnemis, mating success

Abstract

The mating ability ofE. plorans was tested in laboratory conditions in six experimental units composed often males and fifteen females during 31 days. When significant differences were found (three from the six cages, and in totals) they involved a decrease of matings involving males with B chromosomes. The same tendency seems to exist in females, but to a lesser extent, so that a significant effect is only detected when the totals are considered. Accessory chromosomes also delay, in both sexes, the occurrence of the first mating. No mating preferences depending on the number of Bs were detected.

Introduction

The existence of accessory or B chromosome systems has been in many cases explained under the 'para- sitic' model, which claims that Bs are harmful but they are maintained by special mechanisms that increase their transmission to the next generations. However, an important number of unsolved questions about the B systems of many species still exist. An alternative point of view - - the 'heterotic' model - - claims that in some moments of the life cycle or under some cir- cumstances, the individuals carrying a low number of B chromosomes may have a selective advantage over those showing the basic chromosome comple- ment. Hence accessories could be maintained without a special mechanism of accumulation.

Most studies on the accumulation or loss of these chromosomes have traditionally focused on the germ line and particularly on gametogenesis (see the review of Jones & Rees, 1982). The relationship between Bs and fitness has been studied in some plants (see, for instance, Teoh & Jones, 1978; Rees & Hutchinson, 1974; and Puertas, Romera & Pefia, 1985) and in ani- mals like Rattusfuscipes (Thompson, 1984) or Pseu- dococcus obscurus (Nur, 1966a, 1966b, 1969).

In the grasshopper Eyprepocnemis plorans the B chromosome system shows a very high polymorphism for C chromosome structure (Henriques-Gil, Santos & Arana, 1985) and the distribution of different B types among natural populations indicates that newly arisen types replaced the previous ones (Henriques-Gil & Arana, 1990). The main question with this system is the cause for both the presence of Bs in this species and the substitution of one B type by another, since no accumulation mechanisms have been found in E. plorans (L6pez-Le6n et al., 1992a).

Because in animals mating is obviously a critical moment for any kind of genetic transmission, how it is affected by B chromosomes needs detailed analy- sis. L6pez-Le6n et al. (1992b) collected male-female pairs in natural populations ofE. plorans and found no relationship with their B chromosome constitution. In this study we analyse the effect of B chromosomes on mating proficiency under controlled conditions.

Materials and methods

Fertilized females of Eyprepocnemis plorans subsp. plorans were collected from three Spanish natural pop-

198

ulations: Guadalhorce (GH), Fuengirola (FG) and Tor- rox (TX). Each one shows a different predominant type of B chromosome (B1, Bs-B6 and B2 respectively; see Henriques-Gil & Arana, 1990). Pods were obtained in the laboratory and the offspring were simultaneously reared in the conditions described by Henriques-Gil, Santos and Gir~ildez (1982) except that individuals of the same population were raised together in a common cage. The fifth instar hoppers were sexed and separat- ed. All individuals selected for the experiment moulted in a four day interval.

For each of the three populations two different cages (FG1, FG2, GH1, GH2, TX1, and TX2) were established, with 10 males and 15 females per cage, introduced two weeks after the emergence of adults. One male (at FG1) and three females (one at GH2 and two at TX2) died during the first week of observation and were excluded from the analysis. The grasshop- pers were selected from their main population culture by having a normal phenotype (a low number of indi- viduals had some morphological abnormality due to a defective moult and so they were not included in the experiment). All the animals were marked with a num- bered color flag on the femur of each of the jumping legs. The six experimental units were established at the same time and during each of the following 31 days, the observed couples were scored twice a day (morn- ing and evening). True matings were considered those with more than 15 min duration. Because the total time of mating may last several hours, a 'same pair' mat- ing registered in the morning and in the evening was considered a single mating.

Finally, after the period of observation, all the ani- mals were weighed (excluding the third pair of legs since during the experiment, as in natural conditions, one of them was frequently lost). The ovarioles of females previously treated with 0.25% colchicine and male testes without previous treatment were fixed in acetic-ethanol 1 : 3. The preparations of the fixed mate- rial were stained with conventional C-banding. When one or more Bs were present, they corresponded in all cases to the predominant type of their respective geographic area.

Results

The matings scored in the six cages during the 31 days of observation are shown in Table 1. The data are given per day since the numbers obtained in the mornings and the evenings were very similar. It can be observed

that the number of matings increases with time; this is probably the result of the achievement of complete maturity by the grasshoppers.

The total number of matings, however, differs sig- nificantly among populations, being higher for GH and lower for TX. The reasons for such differences remain unknown, but they must reflect some genetic difference since all the animals were obtained and cultured in the same laboratory conditions, developed simultaneous- ly, and the experiment was carried out at the same time. The number of B chromosomes in the males or in the females (both given in Table 2) cannot be responsible for the different behaviour because no correlation was detected with the number of matings.

Matings do not seem to be established in a com- pletely random manner. Indeed, despite the number of males being lower than that of females in a cage (10 and 15), there were a considerable number of males (11 from 59) that were never seen in a mating, while others were involved in a surprisingly high number of mat- ings (like male 6 of FG2 which mated 19 times with 10 different females). The number of females was higher but, by contrast, non-mating females were scarcer (5 from 87); nevertheless, there are also extreme cases, like female 14 of GH1 which mated 20 times with 7 different males.

No significant correlation was detected between the number of matings and weights, though it is important to note that the animals employed were selected by having a completely 'standard' phenotype (see Mate- rials and methods).

The role of B chromosomes Matings per individual. Assuming that accessories have no effect on this variable, the number of mat- ings involving an individual with one or more Bs (+B) should be proportional to the frequency of B carriers in a given cage. The X 2 comparisons with the observed data are given in Table 3.

In three of the six cages there was a significant increase in the matings involving non B ( - B ) males; the same phenomenon may be observed when the totals are considered. By contrast, the proportion + B / - B in the females showed no significant differences with the correspondent observed data; there is, however, a similar tendency towards a lower number of +B matings so that, when the totals are compared, the X 2 value is significant at a 0.05 level.

The mean number of matings per individual was compared by an analysis of variance for paired obser-

199

Table 1. Male/female pairs scored along the 31 days of observation in the six cages established.

Cage: Day FG 1 FG2 GH 1 GH2 TX 1 TX2

1 6/7 4/4 5/6 2 7/15 7/3 9/1 3 9/1 4/3 2/6 8/7 4 5/7 2/11 4/4 9/15 4/55/8 4/9 5 6/5 7/6 2/1410/6 2/75/2 5/9 6 5/15 3/4 2/14 9/10 8/12 7/9 7 10/11 4/4 4/8 8 4/145/10 5/53/8 9 7/6 1/7 10/15 6/14 2/6 3/5

10 7/11 10/1 2/2 8/25/3 10/8 9/10 10/5 3/6 1/10 11 10/10 7/146/13 2/2 1/77/8 9/12 7/123/81/10

7/6 10/8 12 7/1 10/8 6/105/1 8/13 1/6 4/25/14 2/13 13 10/14 6/10 3/3 8/3 4/97/23/8 14 5/4 10/7 6/75/8 10/95/l 7/14 4/6

4/6 5/57/89/1 6/89/13/15 4/12

15 7/11 10/4 6/15/8 2 / 1 5 6/118/144/7 8/113/127/1 3/9 1/2 4/11 10/69/2 2/9

16 1/13 10/3 6/82/1 1/2 i0/11 5/8 4/6 7/65/83/9 4/2 17 5/11 7/155/13 6/9 5/14 2/67/12 4/2 1/78/87/14 1/1 10/12 10/93/2 4/15 18 6/2 1/6 10/127/14 4/7 7/59/65/8

10/7 19 7/8 4/6 7/5 4/116/2 7/89/93/2 3/3 1/12 4/95/5 3/6

2/35/10 20 5/9 1/9 6/13 2/13 5/6 3/89/11 7/5 4/6 10/7 3/6 4/15

2/7 10/1 21 10/145/17/4 2/136/12 9/62/34/77/14 5/2 10/10 4/6 10/129/9 4/13 1/123/15 22 5/91/14 6/114/12/14 6/112/21/124/14 7/159/84/46/11 7/13/2

10/10 10/145/9 23 9/7 5/14 5/910/11 6/8 6/137/22/14 3/129/5 4/15 7/13 1/13 24 10/107/7 4/149/5 6/2 10/76/65/14 I/5 10/5 3/9

10/95/13 25 5/12 10/95/8 2/7 10/9 4 /10 8/77/1410/65/12 4/97/15

6/13 26 4/1 10/149/15 5/149/94/12 8/62/12 1/7 4/6 1/88/15

10/13 3/1 27 5/11 1/107/9 6/25/13 10/1 1/75/68/13 2/14 7/8 4/6

10/8 28 9/7 4/45/11 4/23/4

1/10 29 8/15

3/2 4/6

3/2 4/8

2/4 4/89/14 5/9 6/2 4/4 6/5 4/13 9/14

30 10/25/13 7/32/1 6/2 4/6 10/115/14

31 9/9 4/85/10 10/5 9/12

Total number of matings: 52

6/1410/13 4/11

2/27 /145/8 4/5 9/12 9/46/73/67/7

3/11 6/148/125/9 4/6 7/39/7 1/15 4/119/5 2/1410/7 3/12 1/13

64 86 95 24 48

5/117/14 10/87/5 3/2 10/1 4/3 1/10

9/146/65/67/79/15 7/5 1/159/2 10/9 5/I 8/8 6/68/42/144/8 9/52/146/75/8 3/21/9

200

Table 2. B chromosome constitution of the males and females employed. Male 8 of FGI, female 13 of GH2 and females 3 and 7 of TX2 died in the first week of observation; they were never involved in a mating and were not included in the analysis.

Cage

FG1 FG2 GH1 GH2 TX1 TX2

Male number 1: 2 1 0 2 1 1

2: 2 0 0 1 0 1

3: 0 0 0 0 0 0

4: 2 0 0 0 0 1

5: 0 0 1 0 0 0

6: 1 0 0 0 0 1

7: 0 0 0 0 1 0 8: 1 0 0 0 1

9: 3 0 1 0 0 1

10: 1 1 1 0 1 1

Female number

1: 1 1 1 0 0 1

2: 0 1 0 0 0 1

3: 1 1 1 0 0

4: 0 1 0 0 0 1

5: 1 0 2 0 0 2

6: 0 0 0 0 1 1

7: 0 0 1 0 0

8: 0 1 0 0 1 0

9: 0 0 0 1 0 0

10: 0 0 0 1 0 1

11: 0 0 1 0 1 0

12: 2 0 0 1 1 1

13: 2 0 2 1 0

14: 2 0 1 0 2 1

15: 0 0 0 1 1 1

vations (mean +B/mean - B per cage) with a previous x/(Y + 1) transformation. The B chromosomes pro- duced a significant decrease in the male mean number of matings (F -- 7.026, 1 and 5 d.f., p < 0.05) and had no effect on that of females (F = 2.768); by con- trast, there was a significant cage effect for females (F = 8.033, 5 and 5 d.f., p < 0.05) but not for males (F ---- 0.763).

Mean number of days up to the first mating. It was already mentioned that the number of matings increased consistently along the period of observa- tion, so the moment of occurrence of the first mating must give an idea about how fast or slow the differ-

ent individuals reach complete sexual maturity. The comparisons were made by an analysis of variance for paired observations excluding the individuals that did not mate during the 31 days (their inclusion with an arbitrary value higher than 31 produces similar results). Both in females and in males, the presence of B chro- mosomes delays the first day of mating (F = 17.15, p < 0.01 and F = 11.06, p < 0.05 for males and females respectively). The cage considered also has a significant effect (F = 6.774 and F = 7.758 for males and females respectively).

Weights. The mean weight of individuals with B chro- mosomes showed no differences with respect to that of individuals with the basic karyotype (t = 0.941 and t = 0.164, for males and females respectively). Ana- lysts of variance could not detect differences between populations or between cages.

Assortative mating. A possible tendency of a given type of mating depending on the presence or absence of B chromosomes in males and females was tested by con- tingency X 2, shown in Table 4. The high similarity of observed and expected numbers becomes evident.

Discussion

One of the main features of accessory chromosomes is their lack of effect in the so-called 'exophenotype' of carrier individuals, particularly when qualitative traits are analysed. However, their effect on growth, vigour and fertility is usually towards a reduction of fitness, as predicted by the parasitic model; some exceptions seem to favour the heterotic model (for review, see Jones & Rees, 1982).

The karyotypic analysis of male-female pairs col- lected in the wild favoured the absence of sexual selection based on chromosome number in E. plorans (L6pez-Le6n et al., 1992b). By contrast, the results obtained in this work obviously indicate a harmful effect of B chromosomes. Although their effect is cer- tainly not extreme, there is a clear tendency towards a reduction of the number of matings involving individu- als with Bs and also delaying the occurrence of the first mating. In a study based on established 19 + 2cr trios of E. plorans, L6pez-Le6n et al. (1993) observed a favoured transmission of cytological markers in a sec- ond copulation with respect to the first one. Although such analysis was performed to detect paternity dis- placement, it supports the importance of mating suc-

Table 3. Comparison of the number of matings involving individuals with (+B) and individuals without (--B) accessory chromosomes with the expected (in brackets) from the proportion of both karyotypes in a given cage. All X 2 have 1 degree of freedom. (*: p < 0.05; **: p < 0.01; ***' p < 0.001).

Males Females

Cage Matings +B - B X 2 +B - B X 2

FG1 52 30 (35) 22 (17) 1.91 16 (21) 36 (31) 1.85

FG2 64 7 (19) 57 (45) 11.07"* 21 (21) 43 (43) 0.01

Total 116 37 (55) 79 (61) 11.08"* 37 (43) 79 (73) 1.12

GH1 86 25 (26) 61 (60) 0.04 45 (40) 41 (46) 1.12

GH2 95 11 (19) 84 (76) 4.21" 20 (27) 75 (68) 2.54

Total 181 36 (45) 145 (136) 2.55 65 (69) 116 (112) 0.32

TX1 24 11 (7) 13 (17) 2.87 10 (11) 14 (13) 0.24

TX2 48 24 (34) 24 (14) 9.14"* 28 (33) 20 (15) 2.64

Total 72 35 (36) 37 (36) 0.06 38 (41) 34 (31) 0.54

Total 369 108 (150) 261 (219) 19.91"** 140 (161) 229 (208) 4.94*

201

Table 4. Contingency X 2 tests for assortative mating depending on the number of B chromosomes. Expected numbers in brackets. X 2 was not performed when an expected value was less than 5

Females

Cage Males +B - B X 2 X 2 totals

FG1 +B 11 (9) 19 (21) 1.198

- B 5 (7) 17 (15)

FG2 +B 1 (2) 6 (5)

- B 20 (19) 37 (38)

0.001

GH1 +B 11 (13) 14 (12) 0.978

- B 34 (32) 27 (29)

GH2 +B 3 (2) 8 (9)

- B 17 (18) 67 (66)

0.182

0.222

TX1 +B 4 (5) 7 (6) 0.232

- B 6 (5) 7 (8)

TX2 +B 13 (14) 11 (10) 0.343

- B 15 (14) 9 (10)

0.218

cess: multiple copulations increase the percentage of the sperm contributed and preclude matings by differ- ent males.

The number of matings is more affected for males than for females. A similar phenomenon was observed

by Nur (1966a, 1969) in Pseudococcus where the females were apparently not influenced by the presence of Bs, while male fitness was strongly decreased. If Bs produce a general decrease in mating ability, then such contrast in E. plorans may be expected since the males

202

play a more active role in courtship than the females. The distribution of matings per individual was more uniform in the case of females, while an important number of males did not mate at all, despite being in lower number than the females in the cages.

The delay of the day of first mating also suggests a slower achievement of sexual maturity, Slower devel- opment of grasshoppers with Bs has also been detected for the embryonic stages ofMyrmeleotettix maculatus (Hewitt & East, 1978; Harvey & Hewitt, 1979).

The disparity of results when comparing the dif- ferent cages could be due to different effects of the different Bs, or to an interaction with the genetic back- ground so that accessories would be better tolerated on certain genotypes. Bs are known to differ in cyto- logical behaviour depending on genotypes in cases like Pseudococcus (Nur & Brett, 1985, 1987). In this mate- rial, Nur (1966b) also pointed out a possible interac- tion between Bs and the genetic background for the effect of Bs on fitness. Opposite phenomena seem to occur in plant species like rye (Puertas, Romera & Pefia, 1985). The three populations of Eyprepocnemis employed were chosen because their Bs show differ- ences in size and C-band constitution (detailed pop- ulation analysis on those and other populations was made by Henriques-Gil & Arana, 1990). These, how- ever, may not account for all the contrast, since there are also differences between cages of the same popula- tion. Additionally, the three populations show different behaviour in the number of matings and the mean day of the first mating. In neither case does a relationship with the number of B chromosomes exist.

The effects ofE. plorans accessories could be either due to a general decline in vigour or a specific effect on the ability to mate, rather than a secondary conse- quence of an effect in a particular phenotypic trait. The usual lack of influence on the exophenotype is con- firmed here for animal weight and may be also applied to other features (Camacho et al., 1980).

Accordingly, there are no mating preferences depending on the B chromosomes (see Table 4). This, of course, does not imply that preferences based on some physiological factors do not exist. For instance, in FG2 female number 2 showed five of its six matings with a same male (No. 6); other similar cases may be found in Table 1.

The finding of a negative effect of Bs is not a direct evidence of parasitism: under the parasitic model, an overall lower fitness of B carriers is balanced by a cyto- logical mechanism of accumulation. The latter does not seem to occur in E. plorans (L6pez-Le6n et al., 1992a)

and, in fact, a 'near-neutral' model explaining the B chromosome system of this species has been recently proposed (Camacho et al., 1995).

Acknowledgements

The authors are indebted to Dr. Juan Luis Santos for his excellent assistance and advice. This work has been partially supported by grant PB-880122 from CICYT of Spain.

References

Camacho, J.EM., A.R. Carballo & J. Cabrero, 1980. The B- chromosome system of the grasshopper Eyprepocnemis plorans subsp, plorans (Charpentier). Chromosoma 80: 163-176.

Camacho, J.EM., J. Cabrero, M.D. L6pez-Le6n & M.W. Shaw, 1995. Evolution of a near-neutral B chromosome. 12th Interna- tional Chromosome Conference (in press).

Harvey, A. W. & G. M. Hewitt, 1979. B chromosomes slow devel- opment in a grasshopper. Heredity 42: 397401.

Henriques-Gil, N. & E Arana, 1990. Origin and substitution of B chromosomes in the grasshopper Eyprepocnemis plorans. Evo- lution 44: 747~53.

Henriques-Gil, N., J.L. Santos & R. Gir~ldez, 1982. Genotype- dependent effect of B-chromosomes on chiasma frequency in Eyprepocnemis plorans (Acrididae: Orthoptera), Genetica 59: 223-227.

Henriques-Gil, N., J.L~ Santos & E Arana, 1984. Evolution of a com- plex B-chromosome polymorphism in the grasshopper Eypre- pocnemis plorans. Chromosoma 89: 290-293.

Hewitt, G.M. & T.M. East, 1978. Effects of B chromosomes on development in grasshopper embryos. Heredity 41: 347-356.

Jones, R.N. & H. Rees, 1982. B chromosomes. Academic Press. L6pez-Le6n, M.D., J. Cabrero, J.EM. Camacho, M.I. Cano &

J.L. Santos, 1992. A widespread B chromosome polymorphism maintained without apparent drive. Evolution 46: 529-539.

L6pez-Le6n, M.D., M.C. Pardo, J. Cabrero & J.P.M. Camacho, 1992b. Random mating and absence of sexual selection for B chromosomes in two natural populations of the grasshopper Eyprepocnemis plorans. Heredity 69: 558-561.

L6pez-Le6n, M.D., J. Cabrero, M.C. Pardo, E. Viseras & J.EM. Camacho, 1993. Paternity displacement in the grasshopper Eyprepocnemis plorans. Heredity 71: 539-545.

Nur, U., 1966a. Harmful supernumerary chromosomes in a mealy bug population. Genetics 54: 1225-1238.

Nut, U., 1966b. The effect of surpernumerary chromosomes on the development of mealy bugs. Genetics 54: 1239-1249.

Nur, U., 1969. Harmful B-chromosoms in a mealy bug: additional evidence. Chromosoma 28: 280-297.

Nut, U. & B.L.H. Brett, 1985. Genotypes suppressing meiotic drive of a B chromosome in the mealy bug, Pseudococcus obscurus. Genetics 110: 73-92.

Nur, U. & B.L.H. Brett, 1987. Control of meiotic drive of B chro- mosomes in the mealy bug Pseudoeoeeus affinis (obseurus). Genetics 115: 499-510.

Puertas, M.J., E Romera & A. Pefia, 1985. Comparison of B chro- mosome effects on Secale cereale and Secale vavilovii. Heredity 55: 229-234.

Rees, H. & J. Hutchinson, 1974. Nuclear DNA variation due to B chromosomes. Cold Spring Harbor Symp. Quant. Biol. 38: 175-182.

203

Teoh, S.B. & R.N. Jones, 1978. B chromosome selection and fitness in rye. Heredity 41: 35--48.

Thompson, R.L., 1984. B chromosomes in Rattusfuscipes II. The transmission of B chromosomes to offspring and population studies: support for the 'parasitic' model. Heredity 52: 363- 372.