HELGOI~NDER MEERESUNTERSUCHUNGEN Helgol~inder Meeresunters., 44, 459-474 (1990) Thermal ecotypes of amphi-Atlantic algae. I. Algae of Arctic to cold-temperate distribution (Chaetomorpha melagonium, Devaleraea ramentacea and Phycodrys rubens) .I Novaczek, G. W. Lubbers & A. M. Breeman Department fo Manne ,ygoloiB lacigoloiB Centre, ytisrevinU fo Groningen; P.O. Box ,41 Ni-9750 AA PIaren ,)nG( The Netherlands ABSTRACT: Three species of Arctic to cold-temperate amphi-Atlantic algae, all occurring also in the North Pacific, were tested for growth and/or survival at temperatures of -20 to 30 ~ When isolates from both western and eastern Atlantic shores were tested side-by-side, it was found that thermal ecotypes may occur in such Arctic algae. Chaetomorpha melagonium was the most eurythermal of the 3 species. Isolates of this alga were alike in temperature tolerance and growth rate but Icelandic plants were more sensitive to the lethal temperature of 25 ~ than were more southerly isolates from both east and west. With regard to Devaleraea ramentacea, one Canadian isolate grew extraordinar- ily well at -2 and 0 ~ and all tolerated temperatures 2-3 ~ higher than the lethal limit (18-20 ~ of isolates from Europe. Concerning Phycodrys rubens, both eastern and western isolates died at 20 ~ but European plants tolerated the lethal high temperature longerl were more sensitive to freezing, and attained more rapid growth at optimal temperatures. The intertidal species, .C melagonium and .D ramentacea, both survived freezing at -5 and -20~ at least for short time periods. C. melagonium was more susceptible than .D ramentacea to desiccation. Patterns of thermal tolerance may provide insight into the evolutionary history of seaweed species. INTRODUCTION Algal species of the North Atlantic Ocean may have had diverse evolutionary origins (Liining, 1985). Some 165 my ago, the developing ocean basin opened into the tropical Tethys Sea, and algae presumably migrated from there (Smith et al., 1981). After North America separated from South America roughly 140 my ago (Lamb, 1977), and thereafter up until 3-5 my ago (Briggs, 1987), algae could also have entered from the eastern Pacific. Other algae could have reached the Atlantic from the north. The present Arctic Sea, originally an arm of the Pacific (Smith et aL, 1981), was cut off from the North Pacific by .aignireB yllaitinI linked ot the Tethys and cificaP Oceans by shallow epicontinental ,saes ti retal may have been yllautriv landlocked rof several million years. Between 40 and 55 my BP ,sekarF( 1979), Greenland and Norway separated, allowing the temperate waters and, ,ylbissop a tcnitsid cool-water flora fo the polar sea ot enter the tropical North .citnaltA During the subsequent period fo global cooling (Thunell & Belyea, 1982), there may have been a burst fo speciation and noitcnitxe as northern populations fo Tethyan (cid:14)9 Biologische Anstalt Helgoland, Hamburg 460 I. Novaczek, G. W. Lubbers & A. M. Breeman and Arctic algae either adapted or perished under the stress (van den Hoek, 1984; Lfining, 1985). Species from the North Pacific Ocean could also have migrated into the Arctic and North Atlantic after the disruption of the Bering land bridge perhaps 2-3 my ago. Evidence of interoceanic migration of molluscs at this time has been found in fossil deposits (McKenna, 1983). Over the past 2.5-3 my, recurring glaciations may have caused extinctions on one or both sides of the ocean (van den Hoek, 1984; Lfining, 1985; van den Hoek & Breeman, 1990). Ice fields also separated algal populations occupying eastern and western sides of the Atlantic. For many pan-Arctic species, effective isolation may have occurred only during glaciations. For Arctic and temperate species with poor dispersive ability, popula- tions may have been separated earlier (prior to 01 my )PB by the sinking of the Greenland-Scotland ridge (Thiede & Endholm, 1983). Warm temperate and tropical taxa could have been split up by the initial coohng of the north Atlantic or they may have had sporadic genetic exchange across the northern islands up until the most recent warm period around 20 my BP (Frakes, 1979). Not only have populations of amphi-Atlantic species been separated to various degrees and for various lengths of time but they have also been subject to different temperature regimes owing to patterns of circulation and upwelling (Breeman, 1988, 1990). The more dramatic thermal fluctuations of the western shores may date back to the development of the Labrador Current system, 3 my BP (Berggren & Hollister, 1977). During glacial periods, however, seasonahty was reduced on this coast (CLIMAP-Project Members, 1981). Evidence of upwelling in the eastern Atlantic during the last glaciation has been found (Thiede, 1979; CLIMAP-project members, 1981) and there is no reason to doubt that this process has moderated summer temperatures in the region for a longer time period. Given the genetic isolation and differing environments of many amphi-Atlantic algal populations, one might expect to find that they have diverged in terms of thermal response. A further hypothesis is that ecotypic variation might be more common in species that have totally disjunct distributions, as opposed to pan-Arctic species that have some possibility of genetic interchange. Furthermore, among those species having a southerly distribution, ecotypic variation might be most prevalent in those that have been in the Atlantic Ocean for the longest time, i.e. those of Tethyan origin. To determine to what extent ecotypic variation in thermal response occurs in amphi- Atlantic species, we have brought into culture isolates of species of various distributional types from both sides of the ocean and tested them side-by-side for the ability to survive and grow in different temperatures. In this paper we report on three pan-Arctic species, Chaetomorpha melagonium (Web. et Mohr) Kfitz., Devaleraea ramentacea (L.) Guiry and Phycodrys rubens (L.) Batt. MATERIALS AND METHODS Isolates of all 3 species were collected from both eastern and western shores of the North Atlantic Ocean (Table .)1 Unialgal clones of all isolates were propagated from vegetative tissue or occasionally from spores (see Table )1 at 5 ~ and/or 10~ in long (16 h) days. To test for growth and survival, each isolate was incubated in growth Thermal ecotypes of amphi-Atlantic algae )I( 461 Table .1 Place and year fo collection of algal isolates. )t( = tetrasporophyte, )m( = male, )f( = female Species/Code Location Year ahpromoteahC muinogalem CMASS Woods Hole, Mass., USA 1791 TIRBC ,ffocsoR Brittany, France 1986 CHELG Helgoland, FRG 7891 CICE1 Iceland 7891 CICE2 Iceland 7891 aearelaveD aecatnemar DCAN1 )t( NW Cove, N.S., Canada 1986 DCAN2 )t( Peggys Cove, N.S., Canada 6891 DCAN2 )m( Peggys Cove, N.S., Canada (from tetraspores) 1986 DCAN3 )t( Peggys Cove, N.S., Canada 6891 DJM )m( Jan Mayen Is. 8791 DNORt Tromse, Norway 2891 DNOR2 Troms~, Norway 7891 DICE )t( Iceland 7891 syrdocyhP rubens PCAN1 Wolf Is., Bay of Fundy, 1986 N.B., Canada (from tetraspores) PCAN2 Anglo, P.E.I., Canada 1986 1RUEP Plougarneau, Brittany, France 7891 2RUEP Plougarneau, Brittany, France 7891 3RUEP Plougarneau, Brittany, France 7891 chambers +_( 1-2~ at temperatures ranging from 5 to 30~ and in water baths _+( 0.5~ at temperatures of 0 and -2~ (i.e. -1.8~ in unfrozen seawater at 33%o). All cultures received cool white fluorescent hght of 10, 20 or 40 ~mol m -z s -t. Sterile PIES medium (McLachlan, 1973) was made from 33 007 sahnity, North Sea water. Prior to either growth or survival trials, all material was moved from the stock condition towards the experimental temperature in stages of no more than 5 ~ wk -t, and then trimmed and acclimated at the experimental temperature for at least 5 days before growth trials began. All isolates of any one species were tested concurrently. For each growth trial, 5 healthy plants (Chaetomorpha melagonium) or plant seg- ments (other species) from each clone were incubated separately in sterile plastic Petri dishes (10 cm ~) containing 50 ml of medium and sealed with parafilm. For trials conducted in water baths, the plant material was incubated in test tubes containing 20 ml of medium. Whole juvenile plants of Chaetomorpha melagonium, all generated from spores in culture and at least 2 mm in length, were used for growth trials. In the case of Devaleraea ramentacea, apices 5--10 mm long were used and in the case of Phycodrys rubens, small leaflets were employed. For each species, the average initial size of experimental subjects was similar in all trials. Using a camera lucida on a dissecting microscope, the lengths of the former two species and the surface area of the latter (after flattening under a glass cover slip) were traced at 7 (cid:141) magnification. The tracings were later measured using a computer digitiser to give the length or, in the case of .P rubens, the surface area. Measurements were made at intervals of 2 to 10 days, depending upon the rate of growth, until a straight hne on semi-log paper, indicating steady logarithmic 462 I. Novaczek, G. W. Lubbers & A. M. Breeman increase, could be drawn between at least three consecutive data points. The relative growth rate (Kain, 1987}, expressed as % increase d -1, was computed for the period of exponential growth. Experiments were conducted at 10 and/or 40 ~tmol m -2 s -1 at temperatures of -2, 0, 5, 10, 15, 18, 20, 2'2, and 25~ Separate growth trials on C. melagonium were conducted as above, using 1-cm fragments of young plants incubated in test tubes. In this case, the increase in cell numbers was monitored. Differences among isolates were tested using a posteriori Student-Neuman-Keuls (SNK) and Scheffe multiple range tests, using subprogram "oneway" from the Statistical Package for the Social Sciences (SPSS). To determine lethal limits, whole plants (including holdfasts) were incubated in 500-ml flasks at extreme temperatures for a period of 3 months, to simulate a winter or summer time span. The medium was changed every 2-3 weeks, after being warmed or cooled to the experimental temperature so as to avoid thermal shocks. Temperatures of -2, 0, 18, 20, 23, 25, 27 and 30~ were tested. A photon fluence rate of 10 ~tmol m -2 s -1 was used because of the tendency of algae to be sensitive to light at sublethal tem- peratures. At 0 ~ incubations were also performed in total darkness. In the case of Chaetomorpha melagonium, both juvenile and full-sized adult plants were tested for thermal tolerance. At the end of the test period, plants were returned to stock culture conditions and monitored for up to 3 months for signs of regrowth. Trials were repeated on at least 2 separate occasions. Response to high temperatures was also investigated by two other procedures for Chaetomorpha melagonium and Devaleraea ramentacea. Firstly, growth rates over 6-8 weeks were monitored at sub-lethal high temperatures. Secondly, to test for short-term lethal effects, plant apices, each in a test tube of medium, were incubated in a cryostat +__( 0.5~ for a two-wk period. Tests were performed on 10 replicates at each of the following temperatures: 16, 18, 20, 22, 24, 26, 28 and 30~ Following the test period, plant material was returned to 10 ~ and monitored for regrowth. Resistances to freezing and/or desiccation were also tested. Plant segments were taken from long-term culture at 5 ~ and held at 0 ~ for at least 1 week before being incubated in the dark at -5 and -20 ~ either in ice or in air. Tissues were cooled to -5~ at a rate of 1.1~ rain -1 and to -20~ at a rate of 2.3~ rain -1. For some tests, plants were pre-dried to various extents at 4 ~ otherwise all tissues were fully hydrated. After test periods ranging from 1 hour to 14 days, the plant material was thawed in seawater at 0 ~ and then transferred to 5 or 10 ~ and monitored for regrowth. RESULTS Chaetomorpha melagonium G r o w t h. The two growth experiments, in which either length (Fig. la-c) or cell number (Fig. 1d-i) of isolates CMASS, CBRIT and CHELG were measured, gave similar results. When lengths were measured, CBRIT had significantly higher rates of elongation than the other isolates at 5-15 ~ but no such differences were apparent in the trial in which cells were counted. All isolates grew slowly at -2 and 0 ~ in both long days (Fig. 1a-f) and short days (Fig. lg-i). Maximum rates of growth generally occurred at 10-15 ~ although growth at the sublethal temperature of 22 ~ could be rapid for a short period of Thermal ecotypes of amphi-Atlantic algae )I( 463 .01 :"01 (~ SSAMC CBRff (cid:14)9 CLEHC length length length ~8 T o" & . v ~ o (3 3( => ~> i2 0:5 , , , , , , o3, G3, o , Temperoture , , ,o , ,,, 671~(, , o~, 53, Temperoture (~ Io: (~) SSAMC | TIRBC ((cid:127) GLEHC llec # llec # llec # ~8 T o- T "o T v ~ "o (cid:12)9 i o 3( (3 0 0 5 01 51 20 52 03 53 0 5 0, s, 02 5= ~ 5'~ Temperoture erutorepnreT (~ Temperoture (~ (~ SSAMC (~ TiRBC (~) GLEHC cell # llec # llec # ~8 T -o o" ~c (3 2 (3 => 6~ o ~ 5 5- ; ,~ o? & o'2 5'2 ~ -5 0 , 5 , 01, 51, 20 , 52+ ~ 5'3 -s Temperoture (C) Temperoture (~ Temperoture (~ Fig. 1. Relative growth rates (mean A standard error) of isolates of ahpromoteahC melagonium at various temperatures and at photon fluence rates of 10 ~mol in -2 s -1 (dashed line) and 40 ~mol m -2 s -1 (sohd line), a-c Increase in filament length in 16-h days. d-f Increase in cell number in 16-h days. g-i Increase in cell number in 8-h days. Isolate indicated in upper right hand corner of each graph time. When monitored over weeks, it was found that growth rates at 22 ~ declined steadily, stabilizing near zero after 3-4 weeks (Fig. 2). At 25 ~ growth approached zero within the 5-day acclimation period. In most cases there was no significant difference between growth rates at high and low photon fluence rates. With the exception of CHELG at 15 ~ rates of cell division were also comparable in long- versus short-day conditions. H i g h - t e m p e r a t u r e t o 1 e r a n c e. When tested for 2-wk periods at high tem- peratures, juvenile plants of isolates CMASS, CBRIT and CHELG survived at tern- 464 I. Novaczek, G. W. Lubbers & A. M. Breeman | le h Los: 5+ AcclTmation Time (wk) | TIRSC length .r ~o + o 1 2 3 4 5 s Acclimation Time (wk) ~6 ~4 8 2>-+ o i 2 3 4 5 s 7 Acclimation Time (wk) Fig. .2 Decline in relative growth rates (mean +._ standard error) of isolates of ahpromoteahC muArogalem over time at 22~ in 16-h days. Isolate indicated in upper right hand corner of each graph peratures up to and including 26~ but all died at 28 ~ {Table 2). In trials lasting 3 months, however, both juvenile and mature plants of the above isolates and of isolates CICE1 and CICE2, were damaged or died at 23 and 25 ~ and died after 2-6 weeks at 27 and 30 ~ (Table 3). Only the Icelandic isolates (CICE1 and CICE2) consistently died at 25 ~ a few rephcates of all 3 more southerly isolates survived this temperature (with damage}. Mature plants tended to persist longer at high lethal temperatures than did the juveniles, and the smaller juveniles were the most sensitive. Low-temperature tolerance. All isolates survived 3 months at 0 and -2 ~ even in total darkness {Table 3). The isolates CMASS, CBRIT and CHELG were compared for resistance to desiccation and freezing (Table 4). In all cases, drying out to 70-80 % of initial fresh weight at 4 ~ was in itself lethal. Plants could, however, survive Thermal ecotypes of amphi-Atlantic algae (I) 465 Table 2. Survival of fragments or apices, respectively, of Chaetomorpha melagonium and Devaleraea ramentacea after incubation for 2 weeks at temperatures of 16 to 30 ~ in long days (16 h) at 20 ~mol m -2 s -1. In each case n = 10 % Survival at each temperature (~ 16 18 20 22 24 26 28 30 .C melagonium CMASS 100 100 100 100 100 80 0 0 CBRIT 100 100 100 100 100 100 0 0 CHELG 100 100 100 100 100 100 0 0 .D ramentacea DCAN1 100 100 100 90 0 0 0 0 DCAN2 )t( I00 I00 i00 80 0 0 0 0 DCAN3 100 100 100 80 0 0 0 0 DNOR1 100 100 50 50 0 0 0 0 DNOR2 100 100 30 10 0 0 0 0 DJM 100 100 60 0 0 0 0 0 DICE 100 100 20 20 0 0 0 0 Table 3. Thermal tolerance of whole plants of Chaetomorpha melagonium in trials of 3 months duration. L = long days (16 h), S = short days (8 h), D = dark; ++ = undamaged, + damaged but recovered, +- = damaged or dead in repeated trials, - = dead; ju = juvenile plants, ad --- adult plants, nd = no data Isolate Temperature (~ and Daylength -2S OL OD 20L 23L 25L 27L 30L ad ad ad ju ad ju ad ju , ad ju ad ju ad CMASS ++ ++ ++ ++ ++ - +- +- + . . . . . CBRIT + + + + + + nd + + nd +- + . . . . . . CHELG + + + + + + nd + + +- + +- + . . . . . CICEI ++ ++ ++ ++ ++ - + . . . . . . nd CICE2 ++ ++ ++ ++ ++ - + . . . . . . nd for a week after being quickly frozen in air or in water at -5 ~ Survivorship, after being frozen in water at -20~ was good after 1 hour but dechned to zero within 24 hottrs. If frozen to -20 ~ in air, some replicates of all isolates survived for 4 or 24 hours. No plant survived a week at -20 ~ in any condition. Devaleraea rarnentacea Growth. In an analysis of variance of growth rates of Devaleraea ramentacea, temperature, photon fluence rate and isolate were all highly significant factors (P < 0.001). Significantly higher rates of growth were recorded in long days than in short days at 0 ~ (Fig. 3). Canadian isolates had growth optima at 0 or 5 ~ (Fig. 3), and at least one of these isolates grew significantly faster than all European plants at -2 and 0 ~ In European plants, maximum growth occurred at 5 or 10~ In long-day conditions, an enhancement of growth in higher photon fluence rates was often statistically significant 466 I. Novaczek, G. W. Lubbers & A. M. Breeman Table 4. Survival of fragments of ahpromoteahC muinogalem after different periods of time at -5 and -20 ~ either immersed in water (W) or in air (D). nd = no data etalosI WID Time % Survival (n = 4-8) -5 ~ -20 ~ CMASS D 4 h 75 0 1 d 100 12 7 d 100 0 W I h nd 100 4 h 100 0 1 d 100 0 7 d 100 0 CBRIT D 4 h 100 12 I d nd 25 7 d 100 0 W 1 h nd 100 4 h 50 0 1 d 100 0 7 d 88 0 CHELG D 4 h 100 25 1 d nd 0 7 d 100 0 W 1 h nd 75 4 h I00 12 i d 100 0 7 d 78 0 in Canadian plants but was rarely observed in European plants. For DCAN2, growth varied significantly with temperature and photon fluence rate but not with phase (tetrasporophyte versus male) or daylength. In most cases, growth rates dropped dramati- cally ta 15 ~ and long-term monitoring revealed that ta both 15 and 18 ~ growth rates declined ot zero within 2-6 weeks (Fig. .)4 DCAN3 was particularly tolerant of high temperatures. tI had a yltnacifingis higher rate of growth than lla others after i week at 18 ~ and was llits growing after 6 weeks at this temperature. High-temperature tolerance. Isolates from the eastern and western Atlantic differed consistently in their upper tolerance .stimil In short-term tests (Table ,)2 at least 80 % of replicates of Canadian isolates tolerated 22 ~ rof 2 weeks, while no more than 50 % of European plants tolerated this treatment. Even at 20~ lla the European clones suffered some mortality after 2 weeks. In 3-month slairt (Table ,)5 ti was found that the upper lethal limit rof the European isolates lay between 18 and 20~ whereas Canadian isolates survived with damage at 20 ~ and died at 23 ~ Low-temperature tolerance. Both European and Canadian plants remained healthy at 0 and -2~ in light and at 0~ in darkness rof 3 months (Table .)5 All isolates were damaged but survived after being dried ot as elttil as 60 % of original fresh weight at 4 ~ Survival was good when plants were frozen in water or in air for up to a week at -5 ~ but they died if first dried at 4 ~ (Table 6). Survivorship dechned with time in material frozen to -20 ~ especially if frozen in air. Contrary to the trend at =5 ~ long-term survivorship in air at -20 ~ was better after pre-drying at 4 ~ Thermal ecotypes of amphi-Atlantic algae (I) 467 01 etyhporop$ male T o" s .c ,o (3 =>. "6 -5 0 Temperature 5 10 15 o~2 C) 25 O3 35 -5 0 Temperature 5 10 15 20 (~ 25 O3 lo: (~ 2NACD (~ 1ROND male As T => ', ~" -5 Q (cid:12)9 erutarepmeT 5 , o1, 15 , 2o ~(~i 25 , ~ :~ Temperature ~( IO. ( (cid:127) (~ 2NACD 2ROND T etyhporop$ T 5-3 o ~ o, =, o2 ;2 ~ ~ -s 0 Temperature 5 10 15 2o (~ ~ 3~ 35 Temperature ~( | 3NACD e~hporop$ etyhporEoCID o" (3 => ~7o o_ -s o Temperature s o1 15 2o (~ =s ~ .~ -5 o Temperature 5 IO I~ 2o (~ 25 3o Fig. 3. Relative growth rates (mean +_ standard error) of isolates of Devaleraea ramentacea at various temperatures at photon fluence rates of 10 pmol m -2 s -1 (dashed line) and/or 40 ~'nol rn -2 s -1 (solid line) in 8-h days (triangles) or 16-h days (circles)(cid:12)9 Isolate indicated in upper right hand comer of each graph, a-d Canadian isolates, e-h European isolates 468 I. Novaczek, G. W. Lubbers & A. M. Breeman | | 1NACD MJD eporophyte '=, male .c o o (cid:12)9 (3 >_.2 .5 o 1 2 3 5 8 1 2 3 5 6 Acclimation T~me (wk) Acclimation T~me (wk) | | 2NACD 1ROND male & 2 (3 (3 .5 o Acc1 limation 2 3 T;me 4 5 (wk) 6 7 Acclimation I 2 3 em~T s )kw( 6 7 sporophyte ~ 4 o 1 z 3 4 5 e 7 Acclimation Time (wk) Acclimation "rime (wk) \I | 3NACD ECID T sporophyte "~ etyhporop: ~4 =J (3 o .5 1:12 6" ,, . . . . . . ~ ~ o o Ac1 climation Time (wk) Ac1 climation 2 3 T~me 5 (wk) s 7 Fig. 4. Decline of relative growth rates (mean - standard error) of isolates of Devaleraea ramentacea over time at 15~ (sohd line) and 18~ (dashed line) in 16-h days at 40 ~mol m -2 s-1..Isolate indicated in upper right hand corner of each graph, a-d Canadian isolates, e-h European isolates