Cytotechnology 5: 147-154, 1991. 9 1991 Kluwer Academic Publishers. Printed in the Netherlands.

Characterization of a new continuous cell line from the flood water mosquito, A e d e s v e x a n s C.A. Mazzacano, U.G. Munderloh and T.J. Kurtti D e p a r t m e n t o f Entomology, University o f Minnesota, St. Paul, M N 55108, USA Received 18 June 1990; acceptedin revised form 20 September 1990

K e y words: A e d e s vexans, UM-AVE1, karyotypes

Abstract A new cell line, UM-AVE1, was established from embryos of the mosquito A e d e s vexans. Banding patterns for the isozymes lactate dehydrogenase (LDH), malate dehydrogenase (MDH), isocitrate dehydrogenase (IDH), xanthine dehydrogenase (XDH), and esterases were compared with those of larval A e d e s vexans tissues as well as those of four other mosquito cell lines and one moth cell line. Karyotype analyses confirmed that the dipteran cell lines were not contaminated with lepidopteran cells, because in all mosquito lines the modal number of chromosomes was 6 (= 2n) or 7. Isozyme electrophoresis established a specific profile for each cell line. Two isozymes present in UM-AVE1 (LDH, IDH) were not detected in larvae; this could be a reflection of the different stages used for cell line isolation and enzyme analysis, or lability of sample preparations. It is significant that extracts from UM-AVE1 cells and A e d e s vexans larvae had an identical double band for XDH, while all other cell lines examined exhibited only a single band.

Introduction The flood water mosquito, A e d e s vexans, is the most common pest mosquito in the temperate zones, but it does not play a significant role in the transmission of important diseases. In Europe, A. vexans functions as the primary vector of Tahyna virus (Bunyaviridae), which causes mild symptoms in humans (Grimstad, 1987). In Japan, Korea, Taiwan and China, a subspecies, A e d e s vexans nipponi, transmits Getah virus (Kono, 1988) causing a mild, febrile disease in horses. This mosquito is also a very low-efficiency vector for Japanese encephalitis virus (Burke and Leake, 1988) and Akabane virus (St. George and Standfast, 1989). Although in the U.S., Eastern and

Western equine encephalitis virus (Morris, 1988; Reisen and Monath, 1989) and St. Louis encephalitis virus (Tsai and Mitchell, 1989) have been isolated from field-caught A. vexans, titers were so low that A. vexans is not considered a functional vector for these viruses. It is not clear what causes the low vectorial capacity of this abundant species, as compared with other temperate climate aedine mosquitoes such as Aedes triseriatus. Behavioral (e.g., host preference) or physiological factors (e.g., refractoriness to virus multiplication) might both be involved. In vitro studies on the growth of arboviruses in cells from A. vexans have until now been hampered by the lack o f a cell line, as the one used in previous studies (Pudney et al., 1982)

148 was shown to be of lepidopteran origin (Greene et al., 1972). In this paper, we report the establishment of a cell line from embryonated eggs of a laboratory colony ofA. vexans mosquitoes. The isoenzyme profile of these cells is compared with that of larvae from the same colony, as well as with cell lines from two mosquitoes (Aedes albopictus and Aedes aegypti) and one moth (the corn ear worm, Heliothis zea).

glucose for RU-TAE12A (Munderloh et al., 1982); and 4% FBS, 0.01 mg/ml cholesterol (from bovine cholesterol concentrate; Irvine Scientific, Santa Ana, CA), 25 mM 4-(2-hydroxyethyl)-l-piperazine ethane sulfonic acid (HEPES) for BCIRL-HZAM1 (McIntosh and Ignoffo, 1983). For UM-AVE1, the pH was adjusted to 7 and FBS added to 5%. The L-15B was diluted 2:3 using sterile, purified water (18 Megohm resistivity) (Milli Q Systems, Millipore, Bedford, M.A) for all mosquito cell lines.

Materials and methods

Karyotype analysis Isolation of Aedes vexans cell line UM-AVE1 Aedes vexans adults from a colonized European strain (R. Kuhn, Joh.-Gutenberg-Universit~it, Mainz, Germany, unpublished) were allowed to deposit eggs over night in a pan of moist, sterilized sand. After incubation at 28~ for two days, approximately 50 eggs were surface-sterilized as described previously (Munderloh et al., 1982). They were transferred to 0.2 ml of L-15B medium (Munderloh and Kurtti, 1989) diluted with water to 67% (see below) and supplemented with 20% fetal bovine serum (FBS), 10% tryptose phosphate broth (TPB), and 100 units/ml of penicillin and 100 ~g/ml of streptomycin (GIBCO, Grand Island, New York). The eggs were broken open using the flattened end of a sterile glass rod and the resulting suspension was seeded into a flat-bottom tube (Nunc, Roskilde) in 2 ml of the same medium. The culture was incubated at 31~ and from 4 weeks after initiation, 1 ml of the medium was replaced at weekly intervals until the first subculture was made by 8 weeks.

Cell lines and culture conditions Table 1 lists the cell lines used in this study. Cells were cultured at 31~ in 25 cm 2 flasks in L-15B medium (Munderloh and Kurtti, 1989), pH 7.5, with 50 mM glucose, supplemented with 2% FBS (GIBCO) for ATC-15 (Singh 1967); 4% FBS for line 59 (Peleg 1968) and ASE1V (Kurtti and Munderloh, 1989); 4% FBS, 1% TPB, 30 mM

Chromosome spreads of the cell lines were prepared according to Nichols et al. (1971). Exponentially growing ceils were treated with 0.1 ~tg/ml colcemid (GIBCO) and incubated at 31~ for 2.5 h (line 59, ATC-15); 4 h (UM-AVE1, ASE1V, BCtRL-HZAM1): or 4.5 h (RU-TAE12A), depending on growth rates. Chromosomes in 50 metaphase sets were counted for each line.

Preparation of extracts for isozyme analysis Fourth instar Aedes vexans larvae were chilled on ice and rinsed several times with sterile deionized water. They were transferred to a glass homogenizer and extraction buffer [50 mM Tris base (Fisher Scientific, Springfield, NJ), 1 m M disodium ethylene diamine tetraacetic acid (Na2. EDTA, Fisher), 1% Triton-X 100 (Sigma, St. Louis, MO), pH 7.5] was added at a ratio of 10 p.1 of buffer per larva. Homogenates were centrifuged at 4~ 13,500 xg for 15 min. The supernatant fluid was collected and an equal volume of stabilization buffer [32.5 mM sodium barbital pH 8.6 (Mallinckrodt, St. Louis, MO), 40% glycerol (MaUinckrodt), 0.01% bromphenol blue (Sigma), 0.023 mg/ml phenyl methyl sulfonyl fluoride (Sigma)] was added. Extracts were stored in 50 ~tl aliquots at - 2 0 ~ for a maximum of 8 weeks. Cells were grown to 90-100% confluency, and detached from the growth substrate with a stream of medium from a 14 gauge, 3 inch laboratory

149 Table 1. Cell Lines

Insect tissue of origin

Code

Reference

Aedes vexans embryo Aedes albopictus larvae Aedes aegypti embryo Anopheles stephensi embryo Toxorhynchites amboinensis embryo Heliothis zea ovary

UM-AVEt ATC-15 line 59 ASE1V RU-TAE12A BCIRL-HZAM1

Munderloh,this paper Singh, 1967 Peleg, 1968 Kurtti & Munderloh, 1989 Munderlohet al., 1982 Mclntosh & Ignoffo,1983

cannula (Becton, Dickinson, Lincoln Park, NJ) attached to a 10 cc syringe. They were centrifuged for 5 rain at 275 xg and the supernatant fluid discarded. The cells were resuspended gently in two volumes of extraction buffer and incubated on ice for 30 min. An equal volume of stabilization buffer was added, and the cells were centrifuged at 4~ at 13,500 xg for 15 min. The supernatant fluids were collected and stored in 50 ~tl aliquots at -20~ for a maximum of 4-6 weeks.

Electrophoresis and detection o f isozymes

The enzymes analyzed and conditions for their electrophoretic separation are listed in Table 2. The running buffers used were either Tris-Borate (TB) buffer pH 8.6 (modified from Shaw and Prasad 1970) ([0.1 M Tris base (Fisher), 0.1 M boric acid (Fisher), 1.5 mM Na2EDTA (Fisher)] or Tris-Citrate (TC) buffer pH 7.1 (Corsaro and Fraser, 1987) [0.155 M Tris base (Fisher), 0.043 M citric acid]. Non-denaturing polyacrylamide gels, 1.0 mm thick, were polymerized using 0.05 % ammonium persulfate (BioRad, Richmond, CA) and 0.1% N,N,N',N'-tetramethylethylenediamine (TEMED) (BioRad). They were cast and run in a BioRad Mini-Protean apparatus. Sample volumes of 5-30 ~tl were layered into wells filled with the appropriate running buffer. Gels were stained overnight at 37~ for LDH, MDH, IDH, or XDH according to Shaw and Prasad (1970). Esterase gels were incubated at 12~ in 1X TB buffer (pH 6.7) for 30 min, and

then stained at room temperature for 2-3 h as described by Shaw and Prasad (1970), except that fast garnet GBC salt (Fluka, Ronkonkoma, NY) was used as the color indicator. Following staining, gels were rinsed twice with deionized water and photographed using Polaroid 667 film. They were soaked in a fixative solution (40% methanol, 10% glacial acetic acid, 3% glycerol) according to the manufacturer (BioRad) for a minimum of 24 h and dried in a BioRad Model 483 slab gel dryer. Relative mobilities of isozymes (distance migrated by sample/distance migrated by standard) (Rf values) were calculated using line ATC15 as the standard (Rf = 1.0). in the case of EST, where there were multiple bands present for ATC-15, the strongest staining (lowest mobility) band was used to determine the Rf values of isozymes from the other lines.

Results Establishment o f UM-AVE1

Two months after initiation, the cells in the original culture had grown to cover about 70% of the substrate. They were then diluted into two culture tubes, and antibiotics were omitted from the medium. After the fifth transfer, the FBS concentration was reduced to 5%, and after another two subcultures, TPB was omitted. The ceils are currently in their 15th subculture and are diluted five-fold in fresh medium every 2 weeks. The line is comprised of a highly heterogeneous mixture of cell types (Fig. 1).

150 Table 2. Conditions for isozyme electrophoresis

Enzymea

LDH, MDH

IDH

XDH

EST

Stacking gel %T/%C Buffer/pH

4.25%/0.12% 0.1XTC/6.6

4.5%/0.12% 0.1XTC/6.5

4.25%/0.12% 0.1XTB/6.9

Separating gel %T/%C Buffer/pH

8%/0.21% 0.1XTC/7.1

8%/0.21% 0.1XTC/7.1

8%/0.21% 0.1XTB/8.1

5%/0.13% 1XTB/8.75

Running conditions Buffer/pH Time Voltage Current

1/4XTC/7.2 1.5 h 200V const. 35mA max.

1XTC/7.1 2h 200V const. 35mA max.

1XTB/8.6 1.5 h 200V const. 35mA max.

1XTB/8.75 45 rain 200V max. 35mA const.

aLDH: Lactate Dehydrogenase; MDH: Malate Dehydrogenase; EST: Esterases; IDH: Isocitrate Dehydrogenase; XDH: Xanthine Dehydrogenase

Fig. 1. Phase contrast image ofAedes vexans cell line UM-AVE1. Bar represents 50 ~tm.

151

100 m INI IN IN []

80 m

O

60

O r

40

59 ATC-15 TAE12A ASElV AVE1

O

20 O

~

R

m

3

5

|

9 m

m

.,.,,-, |

|

9

6 7 8 9 12 number of chromosomes

9

1

13

II

o 14

Fig. 2. Karyotype of mosquito cell lines.

Karyology The chromosome complement of each line (Fig. 2) confirmed that the mosquito lines were indeed of dipteran, and BCIRL-HZAM1 of lepidopteran origin. BCIRL,HZAM1 spreads were examined by microscopy but not counted due to the large number of the lepidopteran chromosomes.

showed a large degree of aneuploidy, but the other 2 embryonic cell lines, RU-TAE12A and UM-AVE1, were predominantly diploid. While there were some differences in karyotype between the mosquito lines (Fig. 2) it was not possible to differentiate between these and other dipteran cells on that basis. Isozyme analysis was then used in an attempt to find a characteristic enzyme profile for each cell line.

Isozyme analysis Table 3 lists the Rf values for the isozymes from each cell line and from A. vexans larvae.

Discussion

Karyology The chromosome smears indicated that all mosquito cell lines were dipteran (2n = 6), without contamination from lepidopteran cells. We considered the seventh chromosome present in some lines to be a fragment, although centromerespecific stains were not used. One of the embryo derived cell lines used in this study, ASE1V,

Isozyme analysis UM-AVE1 is a newly isolated cell line from embryonic tissues ofA. vexans. We hoped to find a set of isozyme systems that would enable us to confirm the origin of UM-AVE1, and to establish its separate identity from other dipteran and lepidopteran cell lines maintained in our laboratory. The differences in the isozyme profiles of the mosquito cell lines examined here can not be attributed to culture conditions, because all mosquito lines were grown under almost identical conditions in diluted L-15B with 4-5% FBS and 33-63 mM glucose. Only RU-TAE12A received in addition 1% TPB. It is not known whether the

152 Table 3. Rfb Valuesfor isozymesof cell lines andA. vexans larvae

Cell lines Enzymea

ATC-15

ASEIV

Line59

RU-TAE12A BCIRL-HZAM1 UM-AVE1

A. vex. larvae

LDH MDH IDH XDH

1.0 1.0 1.0 1.0

0.62 1.06 0.93 0.68

n.d. 0.94 0.82 1.09

n.d. 0.94 1.03 1.01

0.98 0.83 0.85

EST

1.0 1.15

1.04 1.35

1.26

0.96 1.28

0.78 0.87

n.d. 1.06 n.d. 1.07 1.34 1.0 1.56

1.24

0.91 0.79 0.41 0.97 1.37 0.61 1.02

1.0

1.13 aSee footnoteto Table 2 for abbreviationsto enzymes;bvaluesaveragedfrom a minimumof 3 runs; n.d.: this isozymenot detectedin multipleruns enzyme profile of cell lines changes with time in vitro. Nevertheless, lepidopteran cell clones did not display any variation in isozyme profiles (Mazzacano, unpublished), or only in one, EST, displaying a multiple banding pattern (Corsaro and Fraser, 1987). In cloned D r o s o p h i l a embryonic cell lines striking differences in the enzyme pattern were only seen in those derived from different, distinct genotypes (Debec, 1974). Their isozyme pattern was more closely related to that of larvae, because most enzymes active in embryos are of maternal origin. However, Debec (1974) did not find activity of tissue specific enzymes in cell lines, with the possible exception of mannokinase which is barely detectable in imaginal discs and cultured cells. The new cell line described here was derived from a colony of A. v e x a n s that had been established in the laboratory for over a decade, and which can be considered genetically homogeneous. Therefore we do not expect to see major changes in the isozyme profile, although this possibility can not be excluded. Because of the heterogeneous nature of the cell population present in line UM-AVE1, clonal analysis of this line might give rise to more homogeneous strains with desirable characteristics, e.g., high virus susceptibility (Corsaro and Fraser, 1987). However, cell clones will become more heterogenous with time (Mazzacano, unpublished) even if not all cell types represented in the parent will reappear.

The enzymes selected for isozyme analysis were LDH (glycolysis), MDH (Krebs' cycle), IDH (Krebs' cycle), XDH (purine metabolism), and EST (mixed function esterases). Glucose-6phosphate dehydrogenase (G6PDH, pentose cycle) activity was present in all lines but contrary to Herrera and Mukherjee (1981) conditions yielding well-resolved bands could not be found. Strong MDH activity was found in all extracts. While multiple banding patterns corresponding to the cytoplasmic and mitochondrial forms of the enzyme might be expected, the staining was too diffuse to be able to distinguish any but the farthest migrating band. A band of lower mobility in the larvae corresponding to that in UM-AVE1 could not be determined for this reason. In contrast, Herrera and Mukherjee (1981) found two multi-banded zones of MDH activity in mosquito cell lines but Tabachnick and K.nudson (1980) were also unable to use MDH as a marker isozyme. Greene and Charney (1971) obtained a single sharp band of MDH activity in A. albop i c t u s (ATC-15) cells, but these results may be suspect as the other mosquito cell lines used in their study were later proven to be moth cells (Greene et al. 1972). LDH was present in strongly staining amounts in ATC-15, ASE1V, and BCIRL-HZAM1, and was less prominent though consistently detectable in UM-AVE1. Using starch gel electrophoresis, Tabachnick and Knudson (1980) re-

153 ported a very smeared LDH profile for dipteran and lepidopteran cell lines, but with polyacrylamide we obtained sharply defined bands. Herrera and Mukherjee (1982) saw 3-4 bands in mosquito lines but as LDH is an inducible enzyme, this contrast to the single bands we found could be a reflection of different tissue culture techniques. Multiple runs of extracts from line 59 and RU-TAE12A did not yield any staining activity. This was also true of extracts from A. vexans larvae. The presence of LDH in UMAVE1 samples and the absence of activity in larval samples could be due to the fact that different tissues and different developmental stages of insects may express different isozymes (Debec, 1974; Fritz and Fallon, 1987). Cells in culture are subject to anoxic conditions (McLimans, 1972) which would favor production of lactate. Thus, LDH activity might be selected for in cells during their establishment in vitro. In contrast, insect tissues are richly supplied with tracheoles that deliver oxygen directly to cells. Extracts stained for IDH showed some lability. Samples from line 59, RU-TAE12A, BCIRLHZAM1, ATC-15 and A. vexans larvae stained regardless of the length of time they had been stored, within the six week limit. ASE1V had to be electrophoresed within four to seven days, and UM-AVE1 within two days of harvest. Brown and Knudson (1980; 1982) did not describe any such problems in cell lines from A. aegypti (ATC-10), A. albopictus (ATC-15), Anopheles stephensi (LSTM-AS-43), and Toxorhynch#es amboinensis (PRU-TA-9, PRU-TA-42). Herrera and Mukherjee (1982) reported some variability in IDH banding patterns, but gave no indication as to whether this was related to sample age. UM-AVE1 cells and Aedes vexans larvae exhibited the same unique staining pattern for XDH, distinct from all other cell lines which showed only a single band. UM-AVE1 and A. vexans larval extracts, however, had 2 bands (Mr = 0.97 and Mr = 1.37, and Mr = 1.07 and 1.34, respectively). This may be a good indication of the origin of this cell line. Tabachnick and Knudson (1980) were unable to use XDH as a marker isozyme because of poor resolution of the bands.

In contrast, the banding patterns of XDH were the most sharply defined of all the isozymes examined in this study. Herrera and Mukherjee (1982) obtained results similar to ours with single, well-resolved bands of similar relative mobility in all the mosquito lines they examined. Although they reported two XDH bands in an A. aegypti cell line (line 59), their relative mobilties differed by only 0.03. The greatest variety of staining patterns was seen for the EST isozymes, ranging from a single band in line 59 to four bands in BCIRL-HZAM1. EST was particularly useful in identifying cell lines in our laboratory because we obtained consistent results with each cell line exhibiting its own unique banding pattern every time. While samples from UM-AVEI cells and larvae both exhibited 2 bands, they corresponded at only one position (Mr of UM-AVE1 extracts -- 0.61 and 1.02 and Mr of larval extracts = 1.0 and 1.56). It was possible to definitely establish the identity of each cell line used in this study, based upon their unique staining patterns for a combination of five isozymes. Although some of the cell lines exhibited identical staining profiles for one or more of the isozymes examined, no two lines had identical staining patterns for all five.

Acknowledgements This is publication no. 18,276 of the Minnesota Agricultural Experiment Station of the University of Minnesota. Supported by State funds to the University of Minnesota and a Public Health Service grant from the National Institute of Health no. AR 37909.

References 1. Brown SE and Knudson DL (1980) Characterization of invertebrate cell lines III. Isozyme analyses employing cellulose-acetateelectrophoresis.In Vitro 16: 829-832. 2. Brown SE and Knudson DL (1982) Characterization of invertebrate cell lines IV. Isozymeanalysesof dipteran and acarine cell lines. In Vitro 18: 347-350.

154 3. Burke DS and Leake CJ (1988) Japanese encephalitis. In: TP Monath (ed.) The Arboviruses: Epidemiology and Ecology Vol. III (pp. 64-92). CRC Press, Boca Raton, Fla. 4. Corsaro BG and Fraser MJ (1987) Characterization of clonal populations of the Heliothis zea cell line IPLBHZ1075. In Vitro Cell. and Dev. Biol. 23: 855-862. 5. Debec A (1974) Isozymic patterns and functional states of in vitro cultured lines of Drosophila melanogaster. Wilhelm Roux's Arch. 174: 1-19. 6. Fritz MA and Fallon AM (1987) Properties of a ribonuclease from Aedes aegypti larvae. Comp. Biochem. Physiol. 88B: 595-601. 7. Greene AE and Charney J (1971) Characterization and identification of insect cell cultures. Curr. Top. Microbiol. Immunol. 55" 51-61. 8. Greene AE, Charney J, Nichols WJ and Coriell LL (1972) Species identity of insect cell lines. In Vitro 7: 43--48. 9. Grimstad PR (1987) California group virus disease. In: TP Monath (ed.) The Arboviruses: Epidemiology and Ecology Vol. II (pp. 99-136). CRC Press, Boca Raton, Fla. 10. Herrera RJ and Mukherjee AB (1982) Electrophoretic characterization and comparison of dehydrogenases from eight permanent cell lines. Comp. Biochem. Physiol. 72B: 359366. 11. Kono Y (1988) Getah virus disease. In: TP Monath (ed.) The Arbovirnses: Epidemiology and Ecology Vol. III (pp. 22-36). CRC Press, Boca Raton, Fla. 12. Kurtti TJ and Munderloh UG (1989) Advances in the definition of culture media for mosquito ceils. In: J Mitsuhashi (ed.) Invertebrate Cell System Applications (pp. 2129). CRC Press, Inc., Boca Raton, Florida. 13. McIntosh AH and Ignoffo CM (1983) Characterization of five cell lines established from species of Heliothis. Appl. Ent. Zool. 18: 262-269. 14. McLimans WF (1972) The gaseous environment of the mammalian cell in culture, In: GH Rothblat and VJ Cristofalo (eds.) Growth, Nutrition and Metabolism of Cells in Culture Vol. I (pp. 137-170). Academic Press, New York, NY. 15. Morris CD (1988) Eastern equine encephalomyelitis. In: TP Monath (ed.) The Arbovirnses: Epidemiology and Ecology Vol. III (pp. 1-24). CRC Press, Boc a Raton, Fla.

16. Mundefloh UG and Kurtti TJ (1989) Formulation of medium for tick cell culture. Exp. Appl. Acarol. 7: 219229. 17. Munderloh UG, Kurtti TJ and Maramorosch K (1982) Anopheles stephensi and Toxorhynchites amboinensis: aseptic rearing of mosquito larvae on cultured cells. J. Parasitol. 68: 1085-1091. 18. Nichols W, Bradt C and Bowne W (1971) Cytogenetic studies on cells in culture from the class Insecta. Curt. Top. Microbiol. Immunol. 55: 61-69. 19. Peleg J (1968) Growth of arboviruses in monolayers from subcultured mosquito embryo cells. Virology 35: 617619. 20. Pudney M, Leake CJ and Buckley SM (1982) Replication of arbovirus in arthropod in vitro systems: an overview. In: K Maramorosch and J Mitsuhashi (eds.) Invertebrate Cell Culture Applications (pp. 159-194). Academic Press, New York. 21. Reisen WK and Monath TP (1989) Western equine encephalomyelitis. In: TP Monath (ed.) The Arboviruses: Epidemiology and Ecology Vol. V (pp. 90-137). CRC Press, Boca Raton, Fla. 22. Shaw CR and Prasad R (1970) Starch gel electrophoresis of enzymes - A compilation of recipes. Biochem. Genet. 4: 297-320. 23. Singh KRP (1967) Cell cultures derived from larvae of Aedes albopictus and A. aegypti. Curt. Sci. 36: 506-509. 24. St. George TD and Standfast HA (1989) Simbu group viruses with teratogenic potential. In: TP Monath (ed.) The Arbovirnses: Epidemiology and Ecology Vol. IV (pp. 146166). CRC Press, Boca Raton, Fla. 25. Tabachaick WJ and Knudson DL (1980) Characterization of invertebrate cell lines II. Isozyme analyses employing starch gel electrophoresis. In Vitro 16: 392-398. 26. Tsai TF and Mitchell CJ (1989) St. Louis encephalitis. In: TP Monath (ed.) The Arboviruses: Epidemiology and Ecology Vol. IV CRC Press, Boca Raton, Fla.

Address for offprints: C.A. Mazzacano, Department of Entomology, University of Minnesota, St. Paul, MN 55108, USA.