- Email: [email protected]
The role of endogenous antifreeze protein enhancers in the hemolymph thermal hysteresis activity of the beetle Dendroides canadensis
Journal of Insect Physiology 48 (2002) 103–111 www.elsevier.com/locate/jinsphys
The role of endogenous antifreeze protein enhancers in the hemolymph thermal hysteresis activity of the beetle Dendroides canadensis John G. Duman b
a,*
, Anthony S. Serianni
b
a Department of Biological Sciences, University of Notre Dame, Notre Dame, IN 46556, USA Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556, USA
Received 23 May 2001; accepted 4 October 2001
Abstract Antifreeze proteins (AFPs) lower the freezing point of water by a non-colligative mechanism, but do not lower the melting point, therefore producing a difference between the freezing and melting points termed thermal hysteresis. Thermal hysteresis activity (THA) of AFPs from overwintering larvae of the beetle Dendroides canadensis is dependent upon AFP concentration and the presence of enhancers of THA which may be either other proteins or low molecular mass enhancers. The purpose of this study was to determine the relative contributions of endogenous enhancers in winter D. canadensis hemolymph. Winter hemolymph collected over four successive winters (1997–1998 to 2000–2001) was tested. The first three of these winters were the warmest on record in this area, while December of the final year was the coldest on record. Protein and low molecular mass enhancers raised hemolymph THA 60–97% and 35–55%, respectively, based on hemolymph with peak THA for each year collected over the four successive winters. However, the hemolymph AFPs were not maximally enhanced since addition of the potent enhancer citrate (at non-physiologically high levels) resulted in large increases in THA. 13NMR showed that glycerol was the only low molecular mass solute present in sufficiently high concentrations in the hemolymph to function as an enhancer. Maximum THA appears to be 苲8.5 °C. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Antifreeze proteins; Dendroides canadensis; Insect antifreeze proteins; Insect cold adaptation
1. Introduction Antifreeze proteins (AFPs) are characterized by their ability to lower the non-equilibrium freezing point of water while not appreciably affecting the melting point, thereby producing a difference between the freezing and melting points which is termed thermal hysteresis (DeVries, 1986). According to the generally accepted adsorption–inhibition mechanism of action (Raymond and DeVries, 1977), the AFPs depress the freezing point by adsorbing onto the surface of ice crystals at preferred growth sites (Raymond and DeVries, 1977; Raymond et al., 1989; Knight et al., 1991), probably by means of hydrogen bonding (DeVries and Cheng, 1992; Sicheri
* Corresponding author. Tel.: +1-219-631-7496; fax: +1-219-6317413. E-mail address: [email protected] (J.G. Duman).
and Yang, 1995). This adsorption forces ice crystal growth to occur between the AFPs in highly curved (high surface free energy) fronts rather than the preferred low curvature (low surface free energy) fronts. Consequently the temperature must be lowered before ice growth can proceed. Hydrophobic and van der Waals interactions may also be involved in binding of some AFPs to ice (Cheng and Merz, 1997; Chao et al., 1997). Antifreeze proteins were first studied in marine teleost fish (DeVries, 1971; Davies and Hew, 1990; DeVries and Cheng, 1992; Fletcher et al., 2001), but have also been identified in several terrestrial arthropods including insects (Duman 1977, 2001), spiders (Duman, 1979a; Husby and Zachariassen, 1980), mites (Block and Duman, 1989; Sjursen and So¨mme, 2000), and centipedes (Tursman et al., 1994; Tursman and Duman, 1995). In addition, AFPs are now known to occur outside the animal kingdom in plants (Urrutia et al., 1992; Griffith et al., 1992a,b; Hon et al., 1995; Smallwood et
0022-1910/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 0 1 ) 0 0 1 5 0 - 0
104
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
al., 1999; Duman et al., 1993), fungi and bacteria (Duman and Olsen, 1993; Sun et al., 1995). In the freeze-avoiding overwintering larvae of the beetle Dendroides canadensis, AFPs function to promote supercooling by inhibiting ice nucleators present in the hemolymph (Olsen and Duman, 1997a) and gut (Olsen and Duman, 1997b), and to prevent inoculative freezing initiated by external ice across the cuticle (Olsen et al., 1998). Thirteen different, but structurally related, AFPs have been identified in D. canadensis (Duman et al., 1998; Andorfer and Duman, 2000; Duman, 2001). These are quite similar to the AFPs from the beetle Tenebrio molitor (Graham et al., 1997; Liou et al. 1999, 2000). The thermal hysteresis activities (THAs) associated with certain insect AFPs are the highest known. The winter hemolymph of AFP-producing insects generally has 3–6 °C of thermal hysteresis as compared to 0.7– 1.5 °C in fish (DeVries and Cheng, 1992), or 0.2–0.5 °C in plants (Duman et al., 1993). For fish and plants these in vivo levels of thermal hysteresis correlate well with the activities produced by appropriate concentrations of purified AFPs. However, while the winter hemolymph of D. canadensis typically has 3–6 °C of thermal hysteresis (Duman, 1980; Olsen and Duman, 1997a), high concentrations of the pure D. canadensis AFPs (DAFPs) in aqueous buffer generate only 2.0–2.5 °C of thermal hysteresis activity. In order to achieve the thermal hysteresis levels seen in winter hemolymph the presence of AFP enhancers is required. The first enhancers identified were certain proteins to which the AFPs bound (Wu and Duman, 1991; Duman et al., 1992). Interestingly, these enhancers had ice nucleating activity, suggesting that the AFPs bind to the ice nucleators, probably at the ice organizing sites, thereby inhibiting them, and in the process the thermal hysteresis activity of the AFPs is enhanced. The mechanism of thermal hysteresis enhancement probably results from the ability of the AFP–protein enhancer complex to block a larger surface area of the seed crystal than the AFP alone, causing the hysteretic freezing point to be further depressed. This explanation of course requires that the DAFPs bind simultaneously to both the protein enhancer (ice nucleator) and ice. Enhancement of THA by specific antibodies to D. canadensis AFPs lends further proof for this mechanism of enhancement (Wu et al., 1991). More recently, several very active low molecular mass THA enhancers were identified (Li et al., 1998a). These apparently function by a different mechanism than the protein enhancers. A potential explanation may be found in the anisotropic interfacial energy of polygonal ice crystals model for AFP function proposed by Wilson (1994). However, it is not clear why certain small solutes (e.g. citrate, glycerol) are effective enhancers while many others are not. The best of these low molecular mass enhancers is citrate, which increased the THA of an aqueous solution of pure D. canadensis AFP from 1.2
to 6.8 °C. More recently, citrate and glycerol were demonstrated to significantly improve the ability of D. canadensis AFPs to inhibit ice nucleators, both hemolymph protein ice nucleators and bacterial ice nucleators (Duman, 2001). This synergistic effect is likely to be important in the supercooling abilities of D. canadensis larvae in winter. However, the minimum concentration of low molecular mass enhancers necessary for significant enhancement is typically 0.25–0.50 M. Obviously, physiological concentrations of citrate are much lower. However, the normal winter hemolymph concentrations of other enhancers such as glycerol and sorbitol may be sufficiently high to function as endogenous enhancers in the hemolymph of D. canadensis (Duman 1979b, 1980). The purpose of the study described here was to identify potential low molecular mass enhancers and to investigate the physiological roles of endogenous protein and low molecular mass solutes in increasing the hemolymph thermal hysteresis activity of D. canadensis larvae.
2. Materials and methods 2.1. Larvae collection Dendroides canadensis larvae were collected from woodlots near South Bend, Indiana. Hemolymph was collected by puncturing the cuticle with a 28-gauge needle in the dorsal midline, and taking the resulting hemolymph into a 10 µl glass capillary tube. Hemolymph pools, consisting of hemolymph from 苲50 larvae, were prepared for each collection date. The hemolymph was then stored at ⫺70 °C until use. 2.2. Antifreeze protein purification D. canadensis AFPs (DAFPs) were purified from hemolymph using size exclusion chromatography (BioRad, P-30) and reverse phase high pressure liquid chromatography (Vydac, C18, 218TP54) as described previously (Li et al., 1998b). 2.3. Thermal hysteresis activity The capillary technique was used to measure thermal hysteresis activity (DeVries, 1986). The sample (苲5 µl) was placed in a sealed capillary tube, a small (苲0.25 mm) seed crystal was spray-frozen in the sample and the capillary was placed into a refrigerated alcohol bath which was equipped with a viewing chamber through which the crystal could be observed with a microscope. The bath temperature was controlled to ±0.01 °C. The temperature was slowly raised (0.02 °C/5 min) until the crystal disappeared. This temperature was taken as the melting point of the sample. Another crystal was sprayfrozen in the sample, the capillary again placed in the
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
chamber, and the bath temperature lowered (0.1 °C/2.5 min) until the crystal grew. The temperature at which the seed crystal began to grow was taken as the hysteretic freezing point. Aqueous solutions which do not contain thermal hysteresis antifreeze proteins exhibit identical freezing and melting points. However, in the presence of these proteins the freezing point is often depressed several degrees below the melting point. This difference between the melting and freezing points is termed thermal hysteresis. Note that with certain insect antifreeze proteins, including D. canadensis, the measured thermal hysteresis activity is inversely dependent on the size of the ice crystal present in the sample when the hysteretic freezing point is determined (Zachariassen and Husby, 1982). This is important when comparing thermal hysteresis activities reported in the literature. A Clifton nanoliter osmometer may be used to measure thermal hysteresis; however, because this technique employs a smaller seed crystal in the sample than does the capillary technique, thermal hysteresis values measured with the nanoliter osmometer may be considerably greater than those determined by the capillary technique on the same sample. All values reported here were determined with the capillary technique. 2.4. Nuclear magnetic resonance, glucose and glycerol assays 13
C NMR was performed on hemolymph in an attempt to identify organic compounds that might be present at sufficiently high concentrations to enhance thermal hysteresis activity. The 13C{1H} NMR spectrum of D. canadensis hemolymph was obtained on a Varian Unity Plus 600 MHz FT-NMR spectrometer operating at 150.84 MHz for 13C and equipped with a 3 mm microprobe. The hemolymph sample (苲150 µl) was transferred to a 3 mm NMR tube, and 2H2O (苲20 µl) was added for spectrometer locking. The spectrum was obtained at 30 °C over a 0–200 ppm spectral window (0.66 s acquisition time, 1.850 s relaxation delay; 2.51 s pulse interval) and data were averaged overnight (苲18 hours; 苲26 000 scans). The resulting FID was treated with a 3 Hz line-broadening prior to Fourier transformation. 13C Chemical shifts are reported in ppm relative to the C1 signal of α-d-mannopyranose (95.5 ppm). The glucose concentration of the hemolymph was determined using a glucoseoxidase assay kit (Sigma). Glycerol concentration was measured with a glycerokinase assay kit (Boehringer Manheim). 2.5. DAFP concentration Hemolymph DAFP concentrations were determined by means of immunoblots (Western blots), using a slot blot apparatus (Bio-Rad). Two microliters of sample
105
were applied. The primary antibody was affinity purified (Bio-Rad Econo-Pac serum IgG purification column) anti-DAFP IgG antibodies from rabbit antiserum. The second antibody was goat anti-rabbit IgG horseradish peroxidase (Bio-Rad). Standard curves were established with purified DAFP-1. 2.6. Rationale of the study The following treatments of D. canadensis hemolymph were conducted to ascertain the relative importance of endogenous high and low molecular mass DAFP enhancers. To determine whether winter D. canadensis hemolymph is optimally enhanced by endogenous enhancers, 1 M citrate (the most potent enhancer that we have identified) was added to the hemolymph and the thermal hysteresis activity measured. (An increase in thermal hysteresis following addition of citrate indicates that the hemolymph DAFPs are not optimally enhanced by endogenous enhancers.) The relative importance of low molecular mass enhancers was measured by dialyzing the hemolymph (3500 Dalton cutoff dialysis membrane, Spectra/Por), freeze-drying the dialyzed hemolymph and then bringing the sample back to the original volume by addition of distilled water. The thermal hysteresis activity of the dialyzed sample was then compared to that of the original hemolymph. To remove DAFPs from the dialyzed hemolymph, affinity-purified (BioRad Econo-Pac serum IgG purification column) polyclonal anti-DAFP IgG antibodies (33%, v/v) and Pansorbin (Staphylococcus aureus cells coated with protein-A, purchased from Calbiochem, binding capacity 2.05 mg human IgG/ml of cell suspension, 33% v/v) were added to the hemolymph. The mixture was incubated at room temperature for 30 min, centrifuged at 7000g for 5 min at 0 °C and the supernatant removed. The thermal hysteresis activity of the supernatant was determined to verify that the DAFPs had been removed. This supernatant (which contained the protein enhancers but not the hemolymph low molecular mass solutes or DAFPs) was then added to pure D. canadensis AFPs 1 and 2 to determine the relative enhancer activity of the hemolymph proteins.
3. Results Table 1 presents the results of experiments designed to determine whether protein and/or low molecular mass DAFP enhancers play a role in determining the thermal hysteresis activity of D. canadensis hemolymph in winter and to what extent the activities of winter hemolymph DAFPs are enhanced. Tests on hemolymph pools that exhibited the highest yearly thermal hysteresis activities that were identified from various collections over three successive winters (February 1998, January 1999, and
106
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
Table 1 Effects of various treatments (see text) on the thermal hysteresis activity (°C) of hemolymph collected from three successive winters. Values are mean thermal hysteresis of multiple measurements on the sample. Numbers in parentheses in line 1 are melting points and DAFP concentrations of the hemolymph pool. Numbers in parentheses in the other lines indicate the change in thermal hysteresis (in °C and percent change) caused by the treatment. Standard deviations of thermal hysteresis measurements range from 0.01 to 0.24 °C Treatment
1. Hemolymph (pool) 2. Hemolymph (#1)+1 M citrate 3. Hemolymph (#1) dialyzed 4. Dialyzed hemolymph (#3)+antiDAFP IgG and Pansorbin 5. Pure DAFP-1 and 2 6. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4) 7. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4)+1 M citrate 8. Comparison of #5 and #7
Thermal hysteresis (°C) February 24, 1998
January 22, 1999
February 23, 2000
3.07 (⫺1.10; 1.6 mg/ml) 5.47 (↑2.40; 78%) 1.84 (↓1.23; 40%) 0
5.30 (⫺3.05; 2.6 mg/ml) 8.50 (↑3.20; 60%) 3.29 (↓2.01; 38%) 0
2.83 (⫺2.04; 2.4 mg/ml) 6.90 (↑4.07; 144%) 1.83 (↓1.00; 35%) 0
1.23 2.06 (↑0.83; 67%)
1.23 2.00 (↑0.77; 62%)
1.53 2.45 (↑0.92; 60%)
5.62 (↑3.56; 173%)
5.13 (↑3.13; 156%)
7.20 (↑4.75; 194%)
(↑4.39; 357%)
(↑3.90; 317%)
(↑5.67; 370%)
February 2000) are shown. Peak winter thermal hysteresis activities over most of the 20 plus years that we have studied D. canadensis are 5–6 °C, although levels of 3–5 °C were sometimes measured. As line 1 of Table 1 shows, only the January 1999 pool had high (5+ °C) levels of hysteresis. The lower thermal hysteresis generally seen during these years may be because these were the warmest winters on record in this area. (However, the January 1999 collection followed a 2-week cold period.) Line 1 (Table 1) also lists the melting points of the hemolymph pool, along with the DAFP concentrations. The normal summer melting point is approximately ⫺0.60 °C, and therefore the lower melting points of all the hemolymph pools shown in Table 1 indicate some concentration of polyols, etc., in winter. However, only the January 1999 sample had a melting point (⫺3.05 °C) comparable to those seen in previous severe winters. (For example, the mean hemolymph melting point of D. canadensis larvae collected in February 1978, the coldest winter on record in this area, was ⫺4.68 °C.) Note that DAFP concentrations in the hemolymph were similar over the three years, although the January 1999 levels were highest. In summer hemolymph DAFP concentrations were too low to be measured using this technique; however, low levels of thermal hysteresis are present. The addition of 1 M citrate (the best enhancer that we have identified, although it is being added in non-physiological concentrations) to the hemolymph (line 2, Table 1) resulted in a considerable increase in thermal hysteresis (60–144%; 2.40 to 4.07 °C) showing that, based on the thermal hysteresis activity of the hemolymph, the DAFPs in the native hemolymph were not optimally enhanced by existing endogenous enhancers. However, dialysis of the native hemolymph (line 3, Table 1) resulted in a 35– 40% (1.00
to 2.01 °C) decrease in thermal hysteresis, demonstrating that endogenous low molecular mass enhancers do play an important role in increasing the thermal hysteresis activity in the hemolymph. It would be instructive to remove the protein enhancers from the hemolymph while at the same time leaving the DAFPs but it is not presently possible to do so. Therefore, the DAFPs were removed from the dialyzed hemolymph using anti-DAFP antibody and Pansorbin (line 4, Table 1), leaving the other hemolymph proteins (including the protein enhancers) and any other solutes with molecular masses greater than 3500. Purified D. canadensis AFPs 1 and 2 (DAFP-1 and 2) of known thermal hysteresis activity (line 5, Table 1) were then added to the dialyzed hemolymph (therefore lacking low molecular mass enhancers) from which the endogenous DAFP had been removed, and the thermal hysteresis activity measured (line 6, Table 1). As shown, addition of the hemolymph proteins (minus DAFP) to the purified DAFP increased thermal hysteresis activity by 60–67% (0.77 to 0.92 °C) showing that endogenous protein enhancers do play an important role in increasing the thermal hysteresis activity of winter hemolymph. However, as shown by the large increase in activity resulting from the addition of 1 M citrate to the protein enhancer activated DAFP (lines 7 and 8, Table 1), the endogenous protein enhancers alone do not optimally enhance the pure DAFP. Previous studies (Duman 1979b, 1980) demonstrated relatively high concentrations of glycerol (0.9 to 1.3 M) and sorbitol (38 to 164 mM) in the hemolymph of overwintering D. canadensis larvae. However, these two solutes accounted for only 苲45% of the osmotic concentration of the hemolymph at that time. The remaining solutes were not identified. In an effort to determine whether other solutes which might act as enhancers were
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
present at high concentrations in the hemolymph, 13C NMR was performed. The natural abundance 150 MHz 13 C{1H} NMR spectrum of hemolymph from larvae collected on February 3, 2000, is shown in Fig. 1. The dominant signals arise from glycerol (C1 and C3, 64.2 ppm; C2 73.7 ppm). Weaker signals are observed for d-glucose; the characteristic pyranose anomeric carbon signals at 97.5 (β-pyranose) and 93.7 (α-pyranose) ppm are clearly observed, in addition to the remaining nonanomeric signals observed between 60 and 80 ppm. A very weak signal is also detected at 94.9 ppm and is consistent with the C1 signal of trehalose. An independent glycerol kinase assay of this hemolymph yielded a glycerol concentration of 503 mM, and a glucose oxidase assay gave a glucose concentration of 24 mM. These data are consistent with the relative intensities of the glycerol and glucose signals in the 13C NMR spectrum. Close inspection of the 60–75 ppm region of the 13 C spectrum revealed the possible presence of sorbitol, but its concentration, if present, would be appreciably lower than that of glucose. As observed previously (Kukal et al., 1988), signals clustered near 苲30, 苲130 and 苲180 ppm are attributed to the saturated, unsaturated and carbonyl carbons of fatty acids and/or their esters. Therefore, the only low molecular mass solute likely to be present in sufficiently high concentration to enhance DAFPs in this D. canadensis hemolymph is glycerol. Except for the January 1999 sample, the hemolymph pools used in the experiments shown in Table 1 had lower thermal hysteresis activity than the 5–6 °C normally seen in midwinter of previous years. However, the values shown in Table 1 (line 1) were the highest activities identified in each of the three winters. For compari-
107
son, Table 2 shows the results of the same manipulations of hemolymph from larvae collected earlier in the winter (December 20, 1999) when the thermal hysteresis activity was somewhat less. The season prior to this collection had been unusually warm and the thermal hysteresis at this time was only 2.29 °C (line 1, Table 2), while the melting point was ⫺1.60 °C indicating some increase in polyol concentration. As expected, addition of 1 M citrate to the hemolymph resulted in a large increase in activity (line 2, Table 2), demonstrating that the DAFPs were not optimally enhanced when the hemolymph was collected. Interestingly, dialysis of the hemolymph to remove low molecular mass enhancers (line 3, Table 2) resulted in a large decrease in activity, to 0.88 °C, much lower than seen with midwinter hemolymph (line 3, Table 1). This might suggest the presence of high levels of low molecular mass enhancers in this hemolymph; however, although the melting point is lower than the normal summer value (approximately ⫺0.60 °C), the melting point (⫺1.60 °C) was not especially low, and the glycerol concentration was 238 mM. Another possible explanation is that this hemolymph had less protein enhancer than was present in the hemolymph samples shown in Table 1, and therefore when the small enhancers were removed by dialysis very little additional enhancer remained. This is verified by line 6 (Table 2) which shows that addition of the hemolymph protein enhancers to purified DAFP-1 and 2 resulted in only a small increase in activity. The data reported in Tables 1 and 2 represent hemolymph from D. canadensis larvae collected during the three warmest winters on record in this area. However, the winter of 2000–2001 was colder, thus affording the
Fig. 1. The natural abundance 13C NMR spectrum of hemolymph from D. canadensis collected on February 3, 2000. The dominant signals arising from glycerol (C1 and C3 at 64.2 ppm and C2 at 73.7 ppm) and glucose signals are labeled. See text for details.
108
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
Table 2 Effects of various treatments (see text) on the thermal hysteresis activity of hemolymph from larvae collected on December 20, 1999. Values are mean thermal hysteresis of multiple measurements on the sample, ± standard deviations. The numbers in parentheses in line 1 are the melting point and DAFP concentration of the sample. Numbers in parentheses in other lines indicate the change in thermal hysteresis (in °C and percent change) caused by the treatment Treatment
Thermal hysteresis (°C)
1. Hemolymph (pool) 2. Hemolymph (#1)+1 M citrate 3. Hemolymph (#1) dialyzed 4. Dialyzed hemolymph (#3)+anti-DAFP IgG and Pansorbin 5. Pure DAFP-1 and 2 6. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4) 7. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4)+1 M citrate 8. Comparison of #5 and #7
2.29±0.17 7.43±0.19 0.88±0.01 0.00 1.60±0.10 1.90±0.05 5.43±0.25
opportunity to compare results obtained during these three warm years with those of a colder one. In fact, December 2000 was the coldest and snowiest December on record in the area. Table 3 presents the results of manipulations of hemolymph (identical to those shown in Tables 1 and 2) from larvae collected in early January 2001. Somewhat surprisingly, the results shown in Table 3 are not much different than those seen in the warmer winters (Table 1). The THA of the hemolymph pool was not especially high (3.31 °C) and the DAFP concentration (1.91 mg/µl) was comparable to those of the three previous winters. Also, the melting point was not particularly low (⫺2.35 °C) indicating that solute concentrations and the glycerol concentration (515 mM) were similar to those of the warmer winters. Likewise, the various manipulations of the hemolymph (Table 3) yielded results similar to those seen previously (Table 1). In certain of the previously described experiments (lines 6 and 7, Tables 1–3) citrate and protein enhancers were added to a fairly low concentration of purified DAFP which alone produced low levels of thermal hysteresis (1.23–1.60 °C). To determine the effects of enhancers when beginning with greater levels of DAFP
(⫺1.60; 2.2 mg/ml) (↑5.14; 224%) (↓1.41; 62%)
(↑0.30; 19%) (↑3.53; 186%)
(↑3.83; 239%)
and thermal hysteresis activity, the following experiments were conducted (Table 4). A higher DAFP concentration was used so that the starting thermal hysteresis, without enhancers, was 2.61 °C (line 1, Table 4). Addition of protein enhancers (from hemolymph of January 2001 larvae) increased activity by 1.09 °C, or 42% (line 3, Table 4). While this is a greater increase in activity in degrees C than was seen upon addition of protein enhancer to the purified DAFP with lesser thermal hysteresis activity (line 6 in Tables 1 and 3 compared with line 3 in Table 4), the percent increase was less, 42% vs. 60 to 97%). By comparison, addition of 1 M citrate increased activity by 5.54 or 212% (line 2, Table 4). Addition of both protein enhancer and citrate simultaneously led to an increase in thermal hysteresis of 5.71 °C, or 219% (line 4, Table 4). This value (8.33 °C) may be near the maximum level of thermal hysteresis to which the DAFPs may be enhanced. Note that the combination of citrate and protein enhancers did not appreciably increase thermal hysteresis activity above that produced by the addition of citrate alone (lines 2 and 5, Table 4).
Table 3 Effects of various treatments (see text) on the thermal hysteresis activity (°C) of hemolymph collected in early January 2001, following the coldest December on record in this area. Values are mean thermal hysteresis of multiple measurements on the sample, ± standard deviations. Numbers in parentheses in line 1 are melting points (°C) and DAFP concentrations (mg/ml) of the hemolymph pool. Numbers in parentheses in other lines indicate the change in thermal hysteresis (in °C and percent change) caused by the treatment Treatment
Thermal hysteresis (°C)
1. Hemolymph (pool) 2. Hemolymph (#1)+1 M citrate 3. Hemolymph (#1) dialyzed 4. Dialyzed hemolymph (#3)+anti-DAFP IgG and Pansorbin 5. Pure DAFP-1 and 2 6. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4) 7. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4)+1 M citrate 8. Comparison of #5 and #7
3.31±0.01 6.55±0.14 1.49±0.01 0.00 1.05±0.16 1.98±0.17 6.03±0.25
(⫺2.35; 1.91 mg/ml) (↑3.24; 98%) (↓1.82; 55%)
(↑1.02; 97%) (↑4.03; 204%)
(↑5.05; 480%)
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
109
Table 4 Effects of various treatments (see text) on the thermal hysteresis activity of an aqueous solution containing purified DAFP-4. Values are mean thermal hysteresis of multiple measurements on the sample, ± standard deviation. Numbers in parentheses indicate the change in thermal hysteresis (in °C and percent changes) caused by the treatment Treatment 1. 2. 3. 4. 5. 6.
Thermal hysteresis (°C)
Purified DAFP 2.61±0.09 Purified DAFP (#1)+1 M citrate 8.15±0.23 (↑5.54; 212%) Purified DAFP (#1)+protein enhancer 3.70±0.12 (↑1.09; 42%) Purified DAFP (#1)+1 M citrate+protein enhancer (compared to #1) 8.33±0.24 (↑5.72; 219%) #4 compared to #2 (↑0.18; 2%) #4 compared to #3 (↑4.63; 125%)
4. Discussion In normal years the thermal hysteresis activity in the hemolymph of D. canadensis larvae from the region of South Bend, Indiana, generally begins to increase from the summer values of 0.5 to 0.7 °C in late September or early October. Activity continues to increase gradually through the autumn, peaks in midwinter, and then slowly decreases in late winter, reaching summer values by some time in April or early May (Duman 1980, 1984). Over the past 25 years, midwinter THA generally peaked at levels between 3 and 6 °C, with activities of 5–6 °C being more common during the late 1970s and 1980s when the winters tended to be colder. However, as seen in Tables 1–3, during the past four winters encompassed by this study (the first three of which were successively the warmest on record in this area), only the January 1999 sampling yielded thermal hysteresis greater than 5 °C. This higher value followed a 2-week cold period in early January. However, the January 2001 collection which followed the coldest December on record in this area produced larvae with only 3.31 °C of THA in the hemolymph. This was unexpected since activities in previous years had appeared to correlate with the severity of the winters and/or cold periods preceding collection. One possible explanation is that although December 2000 was the coldest ever recorded in the area, it also broke all records for snowfall. Therefore, because of the insulation provided by the snow, the temperatures experienced by the larvae in their hibernaculae in partially decomposed logs may not have been especially cold. Another possibility is that the previous series of warm winters did not provide the normal selection pressure for high antifreeze activity. Initially it was thought that these seasonal and yearly variations in thermal hysteresis activity were due solely to changes in concentrations of the antifreeze proteins. However, we now know that DAFP activity can be affected by both protein (Wu and Duman, 1991) and low molecular mass (Li et al., 1998a) enhancers. Thermal hysteresis activity does indeed increase with increasing DAFP concentration (Li et al., 1998a), and northern blots demonstrated seasonal variations in transcription of
dafp-1 (DAFP-1 is the major hemolymph DAFP), which coincides with the seasonal variation in thermal hysteresis (Andorfer and Duman, 2000). However, this study demonstrated that endogenous enhancers, both protein and low molecular mass enhancers (glycerol), significantly affected the thermal hysteresis activities in D. canadensis hemolymph collected over four consecutive winters. Based on 13C NMR of the hemolymph, glycerol was the only low molecular mass solute present in winter hemolymph in sufficiently high concentrations to act as an enhancer. Glycerol, of course, only accumulates in the winter. Protein enhancers likewise appear to vary seasonally. Low levels of protein enhancers seemed to contribute to the low THA measured during the warm December of 1999 (line 6, Table 2), while midwinter levels of enhancement produced by proteins was considerably greater (line 6, Tables 1 and 3). There appear to be multiple protein enhancers present in D. canadensis hemolymph. A 70 kDa enhancer with ice nucleator activity was first purified by Wu and Duman (1991). However, hemolymph lipoproteins and two to three other proteins are also enhancers. In addition, an 苲13 kDa peptide enhancer has also been identified (Duman, unpublished). It is clear that variations in DAFP concentration, along with the levels of protein and glycerol enhancers, are the main factors contributing to seasonal variations in THA. While these same factors certainly affect yearly variations in activity, their relative contributions are perhaps less obvious. The hemolymph collected in January 1999 had the highest THA (5.30 °C) of the samples collected. This sample also had the highest DAFP concentration, but it was only slightly greater than that of the February 2000 hemolymph which had the lowest THA (2.83 °C). Therefore, while DAFP concentration is obviously a primary contributing factor in THA, other factors are critical. The January 1999 sample also exhibited the greatest THA (8.5 °C) after citrate enhancement, suggesting the greatest potential for THA. This may reflect the high DAFP concentration plus the presence of other enhancers. In fact, the melting point of the January 1999 sample was the lowest (⫺3.05 °C) indicating the highest glycerol concentrations. (Unfortunately, glycerol con-
110
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
centrations were not measured on the samples in Table 1.) Also, the January 1999 hemolymph exhibited the greatest decrease in THA following dialysis (2.01 °C), indicating that low molecular mass enhancers played a major role in the high THA. The apparent relative contribution of the protein enhancers to the January 1999 THA seems to be similar to that seen in the other years. Consequently, it is apparent from these studies that DAFP concentration, glycerol, and protein enhancers all play an important role in controling hemolymph THA. The role of glycerol is especially interesting as recent studies show that the presence of glycerol, or other low molecular mass enhancers, is critical to the ability of DAFPs to inhibit ice nucleators (Duman, 2001). However, the large increase in THA upon addition of non-physiologically high levels of citrate clearly indicates that hemolymph DAFPs are not maximally enhanced by the protein and low molecular mass enhancers present in winter hemolymph. Perhaps those rare individual D. canadensis larvae with THA of 8–9 °C (Duman 1979b, 1980) have unusual, as yet unidentified, enhancers not present in most larvae. A concern with the data presented in this study results from the necessity of using pooled hemolymph samples to represent the population. (The various manipulations cannot be conducted on the 2–5 µl hemolymph samples collected from each individual.) While the pools should reflect the average of the population, they do, of course, eliminate individual variation, and at worst the pools could reflect average values of a bimodal population in which none of the individuals exhibits the average value. While this is important to bear in mind, it probably does not reflect the situation in these D. canadensis. However, there certainly is variation in hemolymph thermal hysteresis activity in these populations (Duman 1977, 1980). For example, the mean (± standard deviation) thermal hysteresis activity measured in 47 individuals collected in late winter (February 24, 1995) was 2.32±0.50. Unfortunately, this information on individual variation was not determined on the populations from which the pools were gathered for this study. In summary, both protein and low molecular mass enhancers (glycerol) play important roles in increasing the thermal hysteresis activity in the hemolymph of D. canadensis larvae. Seasonal and annual variations in these enhancers contribute to the variations in antifreeze protein activity common in these insects.
Acknowledgements
This study was supported by National Science Foundation grant IBN98-08376.
References Andorfer, C.A., Duman, J.G., 2000. Isolation and characterization of cDNA clones encoding antifreeze protein of the pyrochroid beetle Dendroides canadensis. Journal of Insect Physiology 46, 365–372. Block, W., Duman, J.G., 1989. Presence of thermal hysteresis producing antifreeze proteins in the Antarctic mite, Alaskozetes antarcticus. Journal of Experimental Zoology 250, 229–231. Chao, H., Houston, M.E., Hodges, R.S., Kay, C.M., Sykes, B.D., 1997. A diminished role for hydrogen bonds in antifreeze binding to ice. Biochemistry 36, 14652–14660. Cheng, A., Merz, K.M., 1997. Ice-binding mechanism of winter flounder antifreeze proteins. Biophysical Journal 73, 2851–2873. Davies, P.L., Hew, C.L., 1990. Biochemistry of fish antifreeze proteins. Federation of American Societies for Experimental Biology Journal 4, 2460–2468. DeVries, A.L., 1971. Glycoproteins as biological antifreeze agents in antarctic fishes. Science 172, 1152–1155. DeVries, A.L., 1986. Antifreeze glycopeptides and peptides: interactions with ice and water. Methods in Enzymology 127, 293–303. DeVries, A.L., Cheng, C.-H.C., 1992. The role of antifreeze glycopeptides and peptides in the survival of cold-water fishes. In: Somero, G.N., Osmond, C.B. (Eds.), Water and Life: Comparative analysis of water relationships at the organismic, cellular and molecular level. Springer-Verlag, Berlin, Heidelberg, pp. 301–315. Duman, J.G., 1977. The role of macromolecular antifreeze in the darkling beetle Meracantha contracta. Journal of Comparative Physiology 115B, 279–286. Duman, J.G., 1979a. Subzero temperature tolerance in spiders: the role of thermal hysteresis factors. Journal of Comparative Physiology 131, 347–352. Duman, J.G., 1979b. Thermal hysteresis factors in overwintering insects. Journal of Insect Physiology 25, 805–810. Duman, J.G., 1980. Factors involved in the overwintering survival of the freeze tolerant beetle Dendroides canadensis. Journal of Comparative Physiology B 136, 53–59. Duman, J.G., 1984. Thermal hysteresis antifreeze proteins in the midgut fluid of overwintering larvae of the beetle Dendroides canadensis. Journal of Experimental Zoology 230, 355–361. Duman, J.G., 2001. Antifreeze and ice nucleator proteins in terrestrial arthropods. Annual Review of Physiology 63, 327–357. Duman, J.G., Olsen, T.M., 1993. Thermal hysteresis protein activity in bacteria, fungi, and phylogenetically diverse plants. Cryobiology 30, 322–328. Duman, J.G., Wu, D.W., Yeung, K.L., 1992. Hemolymph proteins involved in the cold tolerance of terrestrial arthropods: antifreeze and ice nucleator proteins. In: Somero, G.N., Osmond, C.B. (Eds.), Water and Life: Comparative analysis of Water Relationships at the Organismic, Cellular and Molecular Level. Springer-Verlag, Berlin, Heidelberg, pp. 282–300. Duman, J.G., Wu, D.W., Olsen, T.M., Urrutia, M., Tursman, D., 1993. Thermal-hysteresis proteins. Advances in Low-Temperature Biology 2, 131–182. Duman, J.G., Li, N., Verleye, D., Goetz, F.W., Wu, D.W., Andorfer, C.A. et al., 1998. Molecular characterization and sequencing of antifreeze proteins from larvae of the beetle Dendroides canadensis. Journal of Comparative Physiology B. 168, 225–232. Fletcher, G.L., Hew, C.L., Davies, P.L., 2001. Antifreeze proteins of teleost fishes. Annual Review of Physiology 63, 359–390. Graham, L.A., Liow, Y.-C., Walker, V.K., Davies, P.L., 1997. Hyperactive antifreeze protein from beetles. Nature 388, 727–728. Griffith, M., Marentes, E., Ala, P., Yang, D.S.C., 1992a. The role of ice binding proteins in frost tolerance of winter rye. In: Li, P.H., Christersson, L. (Eds.), Advances in Plant Cold Hardiness. CRC Press, Boca Raton, Ann Arbor, Boston, London, pp. 174–184. Griffith, M., Ala, P., Yang, D.S.C., Hon, W.-C., Moffatt, B.A., 1992b.
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
Antifreeze protein produced endogenously in winter rye leaves. Plant Physiology 10, 593–596. Hon, W.C., Griffith, M., Mynarz, A., Kwok, Y.C., Yang, D.S., 1995. Antifreeze proteins in winter rye are similar to pathogenesis-related proteins. Plant Physiology 109, 879–889. Husby, J.A., Zachariassen, K.E., 1980. Antifreeze agents in the body fluid of winter active insects and spiders. Experientia 36, 963–964. Knight, C.A., Cheng, C.C., DeVries, A.L., 1991. Adsorption of αhelical antifreeze peptides on specific ice crystal surface planes. Biophysical Journal 59, 409–418. Kukal, O., Serianni, A.S., Duman, J.G., 1988. Glycerol production in a freeze tolerant arctic insect, Gynaephora groenlandica: an in vivo 13 C NMR study. Journal of Comparative Physiology B 158, 175– 183. Li, N., Andorfer, C.A., Duman, J.G., 1998a. Enhancement of insect antifreeze protein activity by solutes of low molecular mass. Journal of Experimental Biology 201, 2243–2251. Li, N., Chibber, B.A.K., Castellino, F.J., Duman, J.G., 1998b. Mapping of disulfide bridges in antifreeze proteins from overwintering larvae of the beetle Dendroides canadensis. Biochemistry 37, 6343–6350. Liou, Y.-C., Thibault, P., Walker, U.K., Davies, P.L., Graham, L.A., 1999. A complex family of highly heterogeneous and internally repetitive hyperactive antifreeze proteins from the beetle Tenebrio molitor. Biochemistry 38, 11415–11424. Liou, Y.-C., Tocilj, A., Davies, P.L., Jia, Z., 2000. Mimicry of ice structure by surface hydroxyls and water of a β-helix antifreeze protein. Nature 406, 322–324. Olsen, T.M., Duman, J.G., 1997a. Maintenance of the supercooled state in overwintering Pyrochroid beetle larvae Dendroides canadensis: role of hemolymph ice nucleators and antifreeze proteins. Journal of Comparative Physiology B 167, 105–113. Olsen, T.M., Duman, J.G., 1997b. Maintenance of the supercooled state in the gut of overwintering Pyrochroid beetle larvae, Dendroides canadensis: role of gut ice nucleators and antifreeze proteins. Journal of Comparative Physiology B 167, 114–122. Olsen, T.M., Sass, S.J., Li, N., Duman, J.G., 1998. Factors contributing to seasonal increases in inoculative freezing resistance in overwintering fire-colored beetle larvae Dendroides canadensis (Pyrochroidae). Journal of Experimental Biology 201, 1585–1594.
111
Raymond, J.A., DeVries, A.L., 1977. Adsorption inhibition as a mechanism of freezing resistance in polar fishes. Proceedings of the National Academy of Sciences USA 86, 881–885. Raymond, J.A., Wilson, P.W., DeVries, A.L., 1989. Inhibition of growth on nonbasal planes in ice by fish antifreeze. Proceedings of the National Academy of Sciences USA 86, 881–885. Sicheri, F., Yang, D.S.C., 1995. Ice-binding structure and mechanism of an antifreeze protein from winter flounder. Nature 375, 427–431. Sjursen, H., So¨ mme, L., 2000. Seasonal changes in tolerance to cold and desiccation in Phauloppia sp. (Acari, Oribatida) from Finse, Norway. Journal of Insect Physiology 46, 1387–1396. Smallwood, M., Worral, D., Byass, L., Elias, L., Ashford, D., Doucet, C.J., 1999. Isolation and characterization of a novel antifreeze protein from carrot (Daucus carota). Biochemical Journal 340, 385– 391. Sun, X., Griffith, M., Pasternak, J.J., Glick, B.R., 1995. Low temperature growth, freezing survival, and production of antifreeze protein by the plant growth promoting rhizobacterium Pseudomonas putida GR 12-2. Canadian Journal of Microbiology 41, 776–784. Tursman, D., Duman, J.G., Knight, C.A., 1994. Freeze tolerance adaptations in the centipede Lithobius forficatus. Journal of Experimental Zoology 268, 347–353. Tursman, D., Duman, J.G., 1995. Cryoprotective effects of thermal hysteresis protein on survivorship of frozen gut cells from the freeze tolerant centipede Lithobius forficatus. Journal of Experimental Zoology 272, 249–257. Urrutia, M.E., Duman, J.G., Knight, C.A., 1992. Plant thermal hysteresis proteins. Biochimica et Biophysica Acta 1121, 199–206. Wilson, P.W., 1994. A model for thermal hysteresis utilizing the anisotropic interfacial energy of ice crystals. Cryobiology 31, 406–412. Wu, D.W., Duman, J.G., 1991. Activation of antifreeze proteins from the beetle Dendroides canadensis. Journal of Comparative Physiology 161, 279–283. Wu, D.W., Duman, J.D., Xu, L., 1991. Enhancement of insect antifreeze protein activity by antibodies. Biochimica et Biophysica Acta 1076, 416–420. Zachariassen, K.E., Husby, J.A., 1982. Antifreeze effect of thermal hysteresis agents protects highly supercooled insects. Nature 298, 865–867.
The role of endogenous antifreeze protein enhancers in the hemolymph thermal hysteresis activity of the beetle Dendroides canadensis John G. Duman b
a,*
, Anthony S. Serianni
b
a Department of Biological Sciences, University of Notre Dame, Notre Dame, IN 46556, USA Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556, USA
Received 23 May 2001; accepted 4 October 2001
Abstract Antifreeze proteins (AFPs) lower the freezing point of water by a non-colligative mechanism, but do not lower the melting point, therefore producing a difference between the freezing and melting points termed thermal hysteresis. Thermal hysteresis activity (THA) of AFPs from overwintering larvae of the beetle Dendroides canadensis is dependent upon AFP concentration and the presence of enhancers of THA which may be either other proteins or low molecular mass enhancers. The purpose of this study was to determine the relative contributions of endogenous enhancers in winter D. canadensis hemolymph. Winter hemolymph collected over four successive winters (1997–1998 to 2000–2001) was tested. The first three of these winters were the warmest on record in this area, while December of the final year was the coldest on record. Protein and low molecular mass enhancers raised hemolymph THA 60–97% and 35–55%, respectively, based on hemolymph with peak THA for each year collected over the four successive winters. However, the hemolymph AFPs were not maximally enhanced since addition of the potent enhancer citrate (at non-physiologically high levels) resulted in large increases in THA. 13NMR showed that glycerol was the only low molecular mass solute present in sufficiently high concentrations in the hemolymph to function as an enhancer. Maximum THA appears to be 苲8.5 °C. 2002 Elsevier Science Ltd. All rights reserved. Keywords: Antifreeze proteins; Dendroides canadensis; Insect antifreeze proteins; Insect cold adaptation
1. Introduction Antifreeze proteins (AFPs) are characterized by their ability to lower the non-equilibrium freezing point of water while not appreciably affecting the melting point, thereby producing a difference between the freezing and melting points which is termed thermal hysteresis (DeVries, 1986). According to the generally accepted adsorption–inhibition mechanism of action (Raymond and DeVries, 1977), the AFPs depress the freezing point by adsorbing onto the surface of ice crystals at preferred growth sites (Raymond and DeVries, 1977; Raymond et al., 1989; Knight et al., 1991), probably by means of hydrogen bonding (DeVries and Cheng, 1992; Sicheri
* Corresponding author. Tel.: +1-219-631-7496; fax: +1-219-6317413. E-mail address: [email protected] (J.G. Duman).
and Yang, 1995). This adsorption forces ice crystal growth to occur between the AFPs in highly curved (high surface free energy) fronts rather than the preferred low curvature (low surface free energy) fronts. Consequently the temperature must be lowered before ice growth can proceed. Hydrophobic and van der Waals interactions may also be involved in binding of some AFPs to ice (Cheng and Merz, 1997; Chao et al., 1997). Antifreeze proteins were first studied in marine teleost fish (DeVries, 1971; Davies and Hew, 1990; DeVries and Cheng, 1992; Fletcher et al., 2001), but have also been identified in several terrestrial arthropods including insects (Duman 1977, 2001), spiders (Duman, 1979a; Husby and Zachariassen, 1980), mites (Block and Duman, 1989; Sjursen and So¨mme, 2000), and centipedes (Tursman et al., 1994; Tursman and Duman, 1995). In addition, AFPs are now known to occur outside the animal kingdom in plants (Urrutia et al., 1992; Griffith et al., 1992a,b; Hon et al., 1995; Smallwood et
0022-1910/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 0 1 ) 0 0 1 5 0 - 0
104
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
al., 1999; Duman et al., 1993), fungi and bacteria (Duman and Olsen, 1993; Sun et al., 1995). In the freeze-avoiding overwintering larvae of the beetle Dendroides canadensis, AFPs function to promote supercooling by inhibiting ice nucleators present in the hemolymph (Olsen and Duman, 1997a) and gut (Olsen and Duman, 1997b), and to prevent inoculative freezing initiated by external ice across the cuticle (Olsen et al., 1998). Thirteen different, but structurally related, AFPs have been identified in D. canadensis (Duman et al., 1998; Andorfer and Duman, 2000; Duman, 2001). These are quite similar to the AFPs from the beetle Tenebrio molitor (Graham et al., 1997; Liou et al. 1999, 2000). The thermal hysteresis activities (THAs) associated with certain insect AFPs are the highest known. The winter hemolymph of AFP-producing insects generally has 3–6 °C of thermal hysteresis as compared to 0.7– 1.5 °C in fish (DeVries and Cheng, 1992), or 0.2–0.5 °C in plants (Duman et al., 1993). For fish and plants these in vivo levels of thermal hysteresis correlate well with the activities produced by appropriate concentrations of purified AFPs. However, while the winter hemolymph of D. canadensis typically has 3–6 °C of thermal hysteresis (Duman, 1980; Olsen and Duman, 1997a), high concentrations of the pure D. canadensis AFPs (DAFPs) in aqueous buffer generate only 2.0–2.5 °C of thermal hysteresis activity. In order to achieve the thermal hysteresis levels seen in winter hemolymph the presence of AFP enhancers is required. The first enhancers identified were certain proteins to which the AFPs bound (Wu and Duman, 1991; Duman et al., 1992). Interestingly, these enhancers had ice nucleating activity, suggesting that the AFPs bind to the ice nucleators, probably at the ice organizing sites, thereby inhibiting them, and in the process the thermal hysteresis activity of the AFPs is enhanced. The mechanism of thermal hysteresis enhancement probably results from the ability of the AFP–protein enhancer complex to block a larger surface area of the seed crystal than the AFP alone, causing the hysteretic freezing point to be further depressed. This explanation of course requires that the DAFPs bind simultaneously to both the protein enhancer (ice nucleator) and ice. Enhancement of THA by specific antibodies to D. canadensis AFPs lends further proof for this mechanism of enhancement (Wu et al., 1991). More recently, several very active low molecular mass THA enhancers were identified (Li et al., 1998a). These apparently function by a different mechanism than the protein enhancers. A potential explanation may be found in the anisotropic interfacial energy of polygonal ice crystals model for AFP function proposed by Wilson (1994). However, it is not clear why certain small solutes (e.g. citrate, glycerol) are effective enhancers while many others are not. The best of these low molecular mass enhancers is citrate, which increased the THA of an aqueous solution of pure D. canadensis AFP from 1.2
to 6.8 °C. More recently, citrate and glycerol were demonstrated to significantly improve the ability of D. canadensis AFPs to inhibit ice nucleators, both hemolymph protein ice nucleators and bacterial ice nucleators (Duman, 2001). This synergistic effect is likely to be important in the supercooling abilities of D. canadensis larvae in winter. However, the minimum concentration of low molecular mass enhancers necessary for significant enhancement is typically 0.25–0.50 M. Obviously, physiological concentrations of citrate are much lower. However, the normal winter hemolymph concentrations of other enhancers such as glycerol and sorbitol may be sufficiently high to function as endogenous enhancers in the hemolymph of D. canadensis (Duman 1979b, 1980). The purpose of the study described here was to identify potential low molecular mass enhancers and to investigate the physiological roles of endogenous protein and low molecular mass solutes in increasing the hemolymph thermal hysteresis activity of D. canadensis larvae.
2. Materials and methods 2.1. Larvae collection Dendroides canadensis larvae were collected from woodlots near South Bend, Indiana. Hemolymph was collected by puncturing the cuticle with a 28-gauge needle in the dorsal midline, and taking the resulting hemolymph into a 10 µl glass capillary tube. Hemolymph pools, consisting of hemolymph from 苲50 larvae, were prepared for each collection date. The hemolymph was then stored at ⫺70 °C until use. 2.2. Antifreeze protein purification D. canadensis AFPs (DAFPs) were purified from hemolymph using size exclusion chromatography (BioRad, P-30) and reverse phase high pressure liquid chromatography (Vydac, C18, 218TP54) as described previously (Li et al., 1998b). 2.3. Thermal hysteresis activity The capillary technique was used to measure thermal hysteresis activity (DeVries, 1986). The sample (苲5 µl) was placed in a sealed capillary tube, a small (苲0.25 mm) seed crystal was spray-frozen in the sample and the capillary was placed into a refrigerated alcohol bath which was equipped with a viewing chamber through which the crystal could be observed with a microscope. The bath temperature was controlled to ±0.01 °C. The temperature was slowly raised (0.02 °C/5 min) until the crystal disappeared. This temperature was taken as the melting point of the sample. Another crystal was sprayfrozen in the sample, the capillary again placed in the
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
chamber, and the bath temperature lowered (0.1 °C/2.5 min) until the crystal grew. The temperature at which the seed crystal began to grow was taken as the hysteretic freezing point. Aqueous solutions which do not contain thermal hysteresis antifreeze proteins exhibit identical freezing and melting points. However, in the presence of these proteins the freezing point is often depressed several degrees below the melting point. This difference between the melting and freezing points is termed thermal hysteresis. Note that with certain insect antifreeze proteins, including D. canadensis, the measured thermal hysteresis activity is inversely dependent on the size of the ice crystal present in the sample when the hysteretic freezing point is determined (Zachariassen and Husby, 1982). This is important when comparing thermal hysteresis activities reported in the literature. A Clifton nanoliter osmometer may be used to measure thermal hysteresis; however, because this technique employs a smaller seed crystal in the sample than does the capillary technique, thermal hysteresis values measured with the nanoliter osmometer may be considerably greater than those determined by the capillary technique on the same sample. All values reported here were determined with the capillary technique. 2.4. Nuclear magnetic resonance, glucose and glycerol assays 13
C NMR was performed on hemolymph in an attempt to identify organic compounds that might be present at sufficiently high concentrations to enhance thermal hysteresis activity. The 13C{1H} NMR spectrum of D. canadensis hemolymph was obtained on a Varian Unity Plus 600 MHz FT-NMR spectrometer operating at 150.84 MHz for 13C and equipped with a 3 mm microprobe. The hemolymph sample (苲150 µl) was transferred to a 3 mm NMR tube, and 2H2O (苲20 µl) was added for spectrometer locking. The spectrum was obtained at 30 °C over a 0–200 ppm spectral window (0.66 s acquisition time, 1.850 s relaxation delay; 2.51 s pulse interval) and data were averaged overnight (苲18 hours; 苲26 000 scans). The resulting FID was treated with a 3 Hz line-broadening prior to Fourier transformation. 13C Chemical shifts are reported in ppm relative to the C1 signal of α-d-mannopyranose (95.5 ppm). The glucose concentration of the hemolymph was determined using a glucoseoxidase assay kit (Sigma). Glycerol concentration was measured with a glycerokinase assay kit (Boehringer Manheim). 2.5. DAFP concentration Hemolymph DAFP concentrations were determined by means of immunoblots (Western blots), using a slot blot apparatus (Bio-Rad). Two microliters of sample
105
were applied. The primary antibody was affinity purified (Bio-Rad Econo-Pac serum IgG purification column) anti-DAFP IgG antibodies from rabbit antiserum. The second antibody was goat anti-rabbit IgG horseradish peroxidase (Bio-Rad). Standard curves were established with purified DAFP-1. 2.6. Rationale of the study The following treatments of D. canadensis hemolymph were conducted to ascertain the relative importance of endogenous high and low molecular mass DAFP enhancers. To determine whether winter D. canadensis hemolymph is optimally enhanced by endogenous enhancers, 1 M citrate (the most potent enhancer that we have identified) was added to the hemolymph and the thermal hysteresis activity measured. (An increase in thermal hysteresis following addition of citrate indicates that the hemolymph DAFPs are not optimally enhanced by endogenous enhancers.) The relative importance of low molecular mass enhancers was measured by dialyzing the hemolymph (3500 Dalton cutoff dialysis membrane, Spectra/Por), freeze-drying the dialyzed hemolymph and then bringing the sample back to the original volume by addition of distilled water. The thermal hysteresis activity of the dialyzed sample was then compared to that of the original hemolymph. To remove DAFPs from the dialyzed hemolymph, affinity-purified (BioRad Econo-Pac serum IgG purification column) polyclonal anti-DAFP IgG antibodies (33%, v/v) and Pansorbin (Staphylococcus aureus cells coated with protein-A, purchased from Calbiochem, binding capacity 2.05 mg human IgG/ml of cell suspension, 33% v/v) were added to the hemolymph. The mixture was incubated at room temperature for 30 min, centrifuged at 7000g for 5 min at 0 °C and the supernatant removed. The thermal hysteresis activity of the supernatant was determined to verify that the DAFPs had been removed. This supernatant (which contained the protein enhancers but not the hemolymph low molecular mass solutes or DAFPs) was then added to pure D. canadensis AFPs 1 and 2 to determine the relative enhancer activity of the hemolymph proteins.
3. Results Table 1 presents the results of experiments designed to determine whether protein and/or low molecular mass DAFP enhancers play a role in determining the thermal hysteresis activity of D. canadensis hemolymph in winter and to what extent the activities of winter hemolymph DAFPs are enhanced. Tests on hemolymph pools that exhibited the highest yearly thermal hysteresis activities that were identified from various collections over three successive winters (February 1998, January 1999, and
106
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
Table 1 Effects of various treatments (see text) on the thermal hysteresis activity (°C) of hemolymph collected from three successive winters. Values are mean thermal hysteresis of multiple measurements on the sample. Numbers in parentheses in line 1 are melting points and DAFP concentrations of the hemolymph pool. Numbers in parentheses in the other lines indicate the change in thermal hysteresis (in °C and percent change) caused by the treatment. Standard deviations of thermal hysteresis measurements range from 0.01 to 0.24 °C Treatment
1. Hemolymph (pool) 2. Hemolymph (#1)+1 M citrate 3. Hemolymph (#1) dialyzed 4. Dialyzed hemolymph (#3)+antiDAFP IgG and Pansorbin 5. Pure DAFP-1 and 2 6. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4) 7. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4)+1 M citrate 8. Comparison of #5 and #7
Thermal hysteresis (°C) February 24, 1998
January 22, 1999
February 23, 2000
3.07 (⫺1.10; 1.6 mg/ml) 5.47 (↑2.40; 78%) 1.84 (↓1.23; 40%) 0
5.30 (⫺3.05; 2.6 mg/ml) 8.50 (↑3.20; 60%) 3.29 (↓2.01; 38%) 0
2.83 (⫺2.04; 2.4 mg/ml) 6.90 (↑4.07; 144%) 1.83 (↓1.00; 35%) 0
1.23 2.06 (↑0.83; 67%)
1.23 2.00 (↑0.77; 62%)
1.53 2.45 (↑0.92; 60%)
5.62 (↑3.56; 173%)
5.13 (↑3.13; 156%)
7.20 (↑4.75; 194%)
(↑4.39; 357%)
(↑3.90; 317%)
(↑5.67; 370%)
February 2000) are shown. Peak winter thermal hysteresis activities over most of the 20 plus years that we have studied D. canadensis are 5–6 °C, although levels of 3–5 °C were sometimes measured. As line 1 of Table 1 shows, only the January 1999 pool had high (5+ °C) levels of hysteresis. The lower thermal hysteresis generally seen during these years may be because these were the warmest winters on record in this area. (However, the January 1999 collection followed a 2-week cold period.) Line 1 (Table 1) also lists the melting points of the hemolymph pool, along with the DAFP concentrations. The normal summer melting point is approximately ⫺0.60 °C, and therefore the lower melting points of all the hemolymph pools shown in Table 1 indicate some concentration of polyols, etc., in winter. However, only the January 1999 sample had a melting point (⫺3.05 °C) comparable to those seen in previous severe winters. (For example, the mean hemolymph melting point of D. canadensis larvae collected in February 1978, the coldest winter on record in this area, was ⫺4.68 °C.) Note that DAFP concentrations in the hemolymph were similar over the three years, although the January 1999 levels were highest. In summer hemolymph DAFP concentrations were too low to be measured using this technique; however, low levels of thermal hysteresis are present. The addition of 1 M citrate (the best enhancer that we have identified, although it is being added in non-physiological concentrations) to the hemolymph (line 2, Table 1) resulted in a considerable increase in thermal hysteresis (60–144%; 2.40 to 4.07 °C) showing that, based on the thermal hysteresis activity of the hemolymph, the DAFPs in the native hemolymph were not optimally enhanced by existing endogenous enhancers. However, dialysis of the native hemolymph (line 3, Table 1) resulted in a 35– 40% (1.00
to 2.01 °C) decrease in thermal hysteresis, demonstrating that endogenous low molecular mass enhancers do play an important role in increasing the thermal hysteresis activity in the hemolymph. It would be instructive to remove the protein enhancers from the hemolymph while at the same time leaving the DAFPs but it is not presently possible to do so. Therefore, the DAFPs were removed from the dialyzed hemolymph using anti-DAFP antibody and Pansorbin (line 4, Table 1), leaving the other hemolymph proteins (including the protein enhancers) and any other solutes with molecular masses greater than 3500. Purified D. canadensis AFPs 1 and 2 (DAFP-1 and 2) of known thermal hysteresis activity (line 5, Table 1) were then added to the dialyzed hemolymph (therefore lacking low molecular mass enhancers) from which the endogenous DAFP had been removed, and the thermal hysteresis activity measured (line 6, Table 1). As shown, addition of the hemolymph proteins (minus DAFP) to the purified DAFP increased thermal hysteresis activity by 60–67% (0.77 to 0.92 °C) showing that endogenous protein enhancers do play an important role in increasing the thermal hysteresis activity of winter hemolymph. However, as shown by the large increase in activity resulting from the addition of 1 M citrate to the protein enhancer activated DAFP (lines 7 and 8, Table 1), the endogenous protein enhancers alone do not optimally enhance the pure DAFP. Previous studies (Duman 1979b, 1980) demonstrated relatively high concentrations of glycerol (0.9 to 1.3 M) and sorbitol (38 to 164 mM) in the hemolymph of overwintering D. canadensis larvae. However, these two solutes accounted for only 苲45% of the osmotic concentration of the hemolymph at that time. The remaining solutes were not identified. In an effort to determine whether other solutes which might act as enhancers were
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
present at high concentrations in the hemolymph, 13C NMR was performed. The natural abundance 150 MHz 13 C{1H} NMR spectrum of hemolymph from larvae collected on February 3, 2000, is shown in Fig. 1. The dominant signals arise from glycerol (C1 and C3, 64.2 ppm; C2 73.7 ppm). Weaker signals are observed for d-glucose; the characteristic pyranose anomeric carbon signals at 97.5 (β-pyranose) and 93.7 (α-pyranose) ppm are clearly observed, in addition to the remaining nonanomeric signals observed between 60 and 80 ppm. A very weak signal is also detected at 94.9 ppm and is consistent with the C1 signal of trehalose. An independent glycerol kinase assay of this hemolymph yielded a glycerol concentration of 503 mM, and a glucose oxidase assay gave a glucose concentration of 24 mM. These data are consistent with the relative intensities of the glycerol and glucose signals in the 13C NMR spectrum. Close inspection of the 60–75 ppm region of the 13 C spectrum revealed the possible presence of sorbitol, but its concentration, if present, would be appreciably lower than that of glucose. As observed previously (Kukal et al., 1988), signals clustered near 苲30, 苲130 and 苲180 ppm are attributed to the saturated, unsaturated and carbonyl carbons of fatty acids and/or their esters. Therefore, the only low molecular mass solute likely to be present in sufficiently high concentration to enhance DAFPs in this D. canadensis hemolymph is glycerol. Except for the January 1999 sample, the hemolymph pools used in the experiments shown in Table 1 had lower thermal hysteresis activity than the 5–6 °C normally seen in midwinter of previous years. However, the values shown in Table 1 (line 1) were the highest activities identified in each of the three winters. For compari-
107
son, Table 2 shows the results of the same manipulations of hemolymph from larvae collected earlier in the winter (December 20, 1999) when the thermal hysteresis activity was somewhat less. The season prior to this collection had been unusually warm and the thermal hysteresis at this time was only 2.29 °C (line 1, Table 2), while the melting point was ⫺1.60 °C indicating some increase in polyol concentration. As expected, addition of 1 M citrate to the hemolymph resulted in a large increase in activity (line 2, Table 2), demonstrating that the DAFPs were not optimally enhanced when the hemolymph was collected. Interestingly, dialysis of the hemolymph to remove low molecular mass enhancers (line 3, Table 2) resulted in a large decrease in activity, to 0.88 °C, much lower than seen with midwinter hemolymph (line 3, Table 1). This might suggest the presence of high levels of low molecular mass enhancers in this hemolymph; however, although the melting point is lower than the normal summer value (approximately ⫺0.60 °C), the melting point (⫺1.60 °C) was not especially low, and the glycerol concentration was 238 mM. Another possible explanation is that this hemolymph had less protein enhancer than was present in the hemolymph samples shown in Table 1, and therefore when the small enhancers were removed by dialysis very little additional enhancer remained. This is verified by line 6 (Table 2) which shows that addition of the hemolymph protein enhancers to purified DAFP-1 and 2 resulted in only a small increase in activity. The data reported in Tables 1 and 2 represent hemolymph from D. canadensis larvae collected during the three warmest winters on record in this area. However, the winter of 2000–2001 was colder, thus affording the
Fig. 1. The natural abundance 13C NMR spectrum of hemolymph from D. canadensis collected on February 3, 2000. The dominant signals arising from glycerol (C1 and C3 at 64.2 ppm and C2 at 73.7 ppm) and glucose signals are labeled. See text for details.
108
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
Table 2 Effects of various treatments (see text) on the thermal hysteresis activity of hemolymph from larvae collected on December 20, 1999. Values are mean thermal hysteresis of multiple measurements on the sample, ± standard deviations. The numbers in parentheses in line 1 are the melting point and DAFP concentration of the sample. Numbers in parentheses in other lines indicate the change in thermal hysteresis (in °C and percent change) caused by the treatment Treatment
Thermal hysteresis (°C)
1. Hemolymph (pool) 2. Hemolymph (#1)+1 M citrate 3. Hemolymph (#1) dialyzed 4. Dialyzed hemolymph (#3)+anti-DAFP IgG and Pansorbin 5. Pure DAFP-1 and 2 6. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4) 7. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4)+1 M citrate 8. Comparison of #5 and #7
2.29±0.17 7.43±0.19 0.88±0.01 0.00 1.60±0.10 1.90±0.05 5.43±0.25
opportunity to compare results obtained during these three warm years with those of a colder one. In fact, December 2000 was the coldest and snowiest December on record in the area. Table 3 presents the results of manipulations of hemolymph (identical to those shown in Tables 1 and 2) from larvae collected in early January 2001. Somewhat surprisingly, the results shown in Table 3 are not much different than those seen in the warmer winters (Table 1). The THA of the hemolymph pool was not especially high (3.31 °C) and the DAFP concentration (1.91 mg/µl) was comparable to those of the three previous winters. Also, the melting point was not particularly low (⫺2.35 °C) indicating that solute concentrations and the glycerol concentration (515 mM) were similar to those of the warmer winters. Likewise, the various manipulations of the hemolymph (Table 3) yielded results similar to those seen previously (Table 1). In certain of the previously described experiments (lines 6 and 7, Tables 1–3) citrate and protein enhancers were added to a fairly low concentration of purified DAFP which alone produced low levels of thermal hysteresis (1.23–1.60 °C). To determine the effects of enhancers when beginning with greater levels of DAFP
(⫺1.60; 2.2 mg/ml) (↑5.14; 224%) (↓1.41; 62%)
(↑0.30; 19%) (↑3.53; 186%)
(↑3.83; 239%)
and thermal hysteresis activity, the following experiments were conducted (Table 4). A higher DAFP concentration was used so that the starting thermal hysteresis, without enhancers, was 2.61 °C (line 1, Table 4). Addition of protein enhancers (from hemolymph of January 2001 larvae) increased activity by 1.09 °C, or 42% (line 3, Table 4). While this is a greater increase in activity in degrees C than was seen upon addition of protein enhancer to the purified DAFP with lesser thermal hysteresis activity (line 6 in Tables 1 and 3 compared with line 3 in Table 4), the percent increase was less, 42% vs. 60 to 97%). By comparison, addition of 1 M citrate increased activity by 5.54 or 212% (line 2, Table 4). Addition of both protein enhancer and citrate simultaneously led to an increase in thermal hysteresis of 5.71 °C, or 219% (line 4, Table 4). This value (8.33 °C) may be near the maximum level of thermal hysteresis to which the DAFPs may be enhanced. Note that the combination of citrate and protein enhancers did not appreciably increase thermal hysteresis activity above that produced by the addition of citrate alone (lines 2 and 5, Table 4).
Table 3 Effects of various treatments (see text) on the thermal hysteresis activity (°C) of hemolymph collected in early January 2001, following the coldest December on record in this area. Values are mean thermal hysteresis of multiple measurements on the sample, ± standard deviations. Numbers in parentheses in line 1 are melting points (°C) and DAFP concentrations (mg/ml) of the hemolymph pool. Numbers in parentheses in other lines indicate the change in thermal hysteresis (in °C and percent change) caused by the treatment Treatment
Thermal hysteresis (°C)
1. Hemolymph (pool) 2. Hemolymph (#1)+1 M citrate 3. Hemolymph (#1) dialyzed 4. Dialyzed hemolymph (#3)+anti-DAFP IgG and Pansorbin 5. Pure DAFP-1 and 2 6. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4) 7. Pure DAFPs (#5)+dialyzed hemolymph with DAFPs removed (#4)+1 M citrate 8. Comparison of #5 and #7
3.31±0.01 6.55±0.14 1.49±0.01 0.00 1.05±0.16 1.98±0.17 6.03±0.25
(⫺2.35; 1.91 mg/ml) (↑3.24; 98%) (↓1.82; 55%)
(↑1.02; 97%) (↑4.03; 204%)
(↑5.05; 480%)
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
109
Table 4 Effects of various treatments (see text) on the thermal hysteresis activity of an aqueous solution containing purified DAFP-4. Values are mean thermal hysteresis of multiple measurements on the sample, ± standard deviation. Numbers in parentheses indicate the change in thermal hysteresis (in °C and percent changes) caused by the treatment Treatment 1. 2. 3. 4. 5. 6.
Thermal hysteresis (°C)
Purified DAFP 2.61±0.09 Purified DAFP (#1)+1 M citrate 8.15±0.23 (↑5.54; 212%) Purified DAFP (#1)+protein enhancer 3.70±0.12 (↑1.09; 42%) Purified DAFP (#1)+1 M citrate+protein enhancer (compared to #1) 8.33±0.24 (↑5.72; 219%) #4 compared to #2 (↑0.18; 2%) #4 compared to #3 (↑4.63; 125%)
4. Discussion In normal years the thermal hysteresis activity in the hemolymph of D. canadensis larvae from the region of South Bend, Indiana, generally begins to increase from the summer values of 0.5 to 0.7 °C in late September or early October. Activity continues to increase gradually through the autumn, peaks in midwinter, and then slowly decreases in late winter, reaching summer values by some time in April or early May (Duman 1980, 1984). Over the past 25 years, midwinter THA generally peaked at levels between 3 and 6 °C, with activities of 5–6 °C being more common during the late 1970s and 1980s when the winters tended to be colder. However, as seen in Tables 1–3, during the past four winters encompassed by this study (the first three of which were successively the warmest on record in this area), only the January 1999 sampling yielded thermal hysteresis greater than 5 °C. This higher value followed a 2-week cold period in early January. However, the January 2001 collection which followed the coldest December on record in this area produced larvae with only 3.31 °C of THA in the hemolymph. This was unexpected since activities in previous years had appeared to correlate with the severity of the winters and/or cold periods preceding collection. One possible explanation is that although December 2000 was the coldest ever recorded in the area, it also broke all records for snowfall. Therefore, because of the insulation provided by the snow, the temperatures experienced by the larvae in their hibernaculae in partially decomposed logs may not have been especially cold. Another possibility is that the previous series of warm winters did not provide the normal selection pressure for high antifreeze activity. Initially it was thought that these seasonal and yearly variations in thermal hysteresis activity were due solely to changes in concentrations of the antifreeze proteins. However, we now know that DAFP activity can be affected by both protein (Wu and Duman, 1991) and low molecular mass (Li et al., 1998a) enhancers. Thermal hysteresis activity does indeed increase with increasing DAFP concentration (Li et al., 1998a), and northern blots demonstrated seasonal variations in transcription of
dafp-1 (DAFP-1 is the major hemolymph DAFP), which coincides with the seasonal variation in thermal hysteresis (Andorfer and Duman, 2000). However, this study demonstrated that endogenous enhancers, both protein and low molecular mass enhancers (glycerol), significantly affected the thermal hysteresis activities in D. canadensis hemolymph collected over four consecutive winters. Based on 13C NMR of the hemolymph, glycerol was the only low molecular mass solute present in winter hemolymph in sufficiently high concentrations to act as an enhancer. Glycerol, of course, only accumulates in the winter. Protein enhancers likewise appear to vary seasonally. Low levels of protein enhancers seemed to contribute to the low THA measured during the warm December of 1999 (line 6, Table 2), while midwinter levels of enhancement produced by proteins was considerably greater (line 6, Tables 1 and 3). There appear to be multiple protein enhancers present in D. canadensis hemolymph. A 70 kDa enhancer with ice nucleator activity was first purified by Wu and Duman (1991). However, hemolymph lipoproteins and two to three other proteins are also enhancers. In addition, an 苲13 kDa peptide enhancer has also been identified (Duman, unpublished). It is clear that variations in DAFP concentration, along with the levels of protein and glycerol enhancers, are the main factors contributing to seasonal variations in THA. While these same factors certainly affect yearly variations in activity, their relative contributions are perhaps less obvious. The hemolymph collected in January 1999 had the highest THA (5.30 °C) of the samples collected. This sample also had the highest DAFP concentration, but it was only slightly greater than that of the February 2000 hemolymph which had the lowest THA (2.83 °C). Therefore, while DAFP concentration is obviously a primary contributing factor in THA, other factors are critical. The January 1999 sample also exhibited the greatest THA (8.5 °C) after citrate enhancement, suggesting the greatest potential for THA. This may reflect the high DAFP concentration plus the presence of other enhancers. In fact, the melting point of the January 1999 sample was the lowest (⫺3.05 °C) indicating the highest glycerol concentrations. (Unfortunately, glycerol con-
110
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
centrations were not measured on the samples in Table 1.) Also, the January 1999 hemolymph exhibited the greatest decrease in THA following dialysis (2.01 °C), indicating that low molecular mass enhancers played a major role in the high THA. The apparent relative contribution of the protein enhancers to the January 1999 THA seems to be similar to that seen in the other years. Consequently, it is apparent from these studies that DAFP concentration, glycerol, and protein enhancers all play an important role in controling hemolymph THA. The role of glycerol is especially interesting as recent studies show that the presence of glycerol, or other low molecular mass enhancers, is critical to the ability of DAFPs to inhibit ice nucleators (Duman, 2001). However, the large increase in THA upon addition of non-physiologically high levels of citrate clearly indicates that hemolymph DAFPs are not maximally enhanced by the protein and low molecular mass enhancers present in winter hemolymph. Perhaps those rare individual D. canadensis larvae with THA of 8–9 °C (Duman 1979b, 1980) have unusual, as yet unidentified, enhancers not present in most larvae. A concern with the data presented in this study results from the necessity of using pooled hemolymph samples to represent the population. (The various manipulations cannot be conducted on the 2–5 µl hemolymph samples collected from each individual.) While the pools should reflect the average of the population, they do, of course, eliminate individual variation, and at worst the pools could reflect average values of a bimodal population in which none of the individuals exhibits the average value. While this is important to bear in mind, it probably does not reflect the situation in these D. canadensis. However, there certainly is variation in hemolymph thermal hysteresis activity in these populations (Duman 1977, 1980). For example, the mean (± standard deviation) thermal hysteresis activity measured in 47 individuals collected in late winter (February 24, 1995) was 2.32±0.50. Unfortunately, this information on individual variation was not determined on the populations from which the pools were gathered for this study. In summary, both protein and low molecular mass enhancers (glycerol) play important roles in increasing the thermal hysteresis activity in the hemolymph of D. canadensis larvae. Seasonal and annual variations in these enhancers contribute to the variations in antifreeze protein activity common in these insects.
Acknowledgements
This study was supported by National Science Foundation grant IBN98-08376.
References Andorfer, C.A., Duman, J.G., 2000. Isolation and characterization of cDNA clones encoding antifreeze protein of the pyrochroid beetle Dendroides canadensis. Journal of Insect Physiology 46, 365–372. Block, W., Duman, J.G., 1989. Presence of thermal hysteresis producing antifreeze proteins in the Antarctic mite, Alaskozetes antarcticus. Journal of Experimental Zoology 250, 229–231. Chao, H., Houston, M.E., Hodges, R.S., Kay, C.M., Sykes, B.D., 1997. A diminished role for hydrogen bonds in antifreeze binding to ice. Biochemistry 36, 14652–14660. Cheng, A., Merz, K.M., 1997. Ice-binding mechanism of winter flounder antifreeze proteins. Biophysical Journal 73, 2851–2873. Davies, P.L., Hew, C.L., 1990. Biochemistry of fish antifreeze proteins. Federation of American Societies for Experimental Biology Journal 4, 2460–2468. DeVries, A.L., 1971. Glycoproteins as biological antifreeze agents in antarctic fishes. Science 172, 1152–1155. DeVries, A.L., 1986. Antifreeze glycopeptides and peptides: interactions with ice and water. Methods in Enzymology 127, 293–303. DeVries, A.L., Cheng, C.-H.C., 1992. The role of antifreeze glycopeptides and peptides in the survival of cold-water fishes. In: Somero, G.N., Osmond, C.B. (Eds.), Water and Life: Comparative analysis of water relationships at the organismic, cellular and molecular level. Springer-Verlag, Berlin, Heidelberg, pp. 301–315. Duman, J.G., 1977. The role of macromolecular antifreeze in the darkling beetle Meracantha contracta. Journal of Comparative Physiology 115B, 279–286. Duman, J.G., 1979a. Subzero temperature tolerance in spiders: the role of thermal hysteresis factors. Journal of Comparative Physiology 131, 347–352. Duman, J.G., 1979b. Thermal hysteresis factors in overwintering insects. Journal of Insect Physiology 25, 805–810. Duman, J.G., 1980. Factors involved in the overwintering survival of the freeze tolerant beetle Dendroides canadensis. Journal of Comparative Physiology B 136, 53–59. Duman, J.G., 1984. Thermal hysteresis antifreeze proteins in the midgut fluid of overwintering larvae of the beetle Dendroides canadensis. Journal of Experimental Zoology 230, 355–361. Duman, J.G., 2001. Antifreeze and ice nucleator proteins in terrestrial arthropods. Annual Review of Physiology 63, 327–357. Duman, J.G., Olsen, T.M., 1993. Thermal hysteresis protein activity in bacteria, fungi, and phylogenetically diverse plants. Cryobiology 30, 322–328. Duman, J.G., Wu, D.W., Yeung, K.L., 1992. Hemolymph proteins involved in the cold tolerance of terrestrial arthropods: antifreeze and ice nucleator proteins. In: Somero, G.N., Osmond, C.B. (Eds.), Water and Life: Comparative analysis of Water Relationships at the Organismic, Cellular and Molecular Level. Springer-Verlag, Berlin, Heidelberg, pp. 282–300. Duman, J.G., Wu, D.W., Olsen, T.M., Urrutia, M., Tursman, D., 1993. Thermal-hysteresis proteins. Advances in Low-Temperature Biology 2, 131–182. Duman, J.G., Li, N., Verleye, D., Goetz, F.W., Wu, D.W., Andorfer, C.A. et al., 1998. Molecular characterization and sequencing of antifreeze proteins from larvae of the beetle Dendroides canadensis. Journal of Comparative Physiology B. 168, 225–232. Fletcher, G.L., Hew, C.L., Davies, P.L., 2001. Antifreeze proteins of teleost fishes. Annual Review of Physiology 63, 359–390. Graham, L.A., Liow, Y.-C., Walker, V.K., Davies, P.L., 1997. Hyperactive antifreeze protein from beetles. Nature 388, 727–728. Griffith, M., Marentes, E., Ala, P., Yang, D.S.C., 1992a. The role of ice binding proteins in frost tolerance of winter rye. In: Li, P.H., Christersson, L. (Eds.), Advances in Plant Cold Hardiness. CRC Press, Boca Raton, Ann Arbor, Boston, London, pp. 174–184. Griffith, M., Ala, P., Yang, D.S.C., Hon, W.-C., Moffatt, B.A., 1992b.
J.G. Duman, A.S. Serianni / Journal of Insect Physiology 48 (2002) 103–111
Antifreeze protein produced endogenously in winter rye leaves. Plant Physiology 10, 593–596. Hon, W.C., Griffith, M., Mynarz, A., Kwok, Y.C., Yang, D.S., 1995. Antifreeze proteins in winter rye are similar to pathogenesis-related proteins. Plant Physiology 109, 879–889. Husby, J.A., Zachariassen, K.E., 1980. Antifreeze agents in the body fluid of winter active insects and spiders. Experientia 36, 963–964. Knight, C.A., Cheng, C.C., DeVries, A.L., 1991. Adsorption of αhelical antifreeze peptides on specific ice crystal surface planes. Biophysical Journal 59, 409–418. Kukal, O., Serianni, A.S., Duman, J.G., 1988. Glycerol production in a freeze tolerant arctic insect, Gynaephora groenlandica: an in vivo 13 C NMR study. Journal of Comparative Physiology B 158, 175– 183. Li, N., Andorfer, C.A., Duman, J.G., 1998a. Enhancement of insect antifreeze protein activity by solutes of low molecular mass. Journal of Experimental Biology 201, 2243–2251. Li, N., Chibber, B.A.K., Castellino, F.J., Duman, J.G., 1998b. Mapping of disulfide bridges in antifreeze proteins from overwintering larvae of the beetle Dendroides canadensis. Biochemistry 37, 6343–6350. Liou, Y.-C., Thibault, P., Walker, U.K., Davies, P.L., Graham, L.A., 1999. A complex family of highly heterogeneous and internally repetitive hyperactive antifreeze proteins from the beetle Tenebrio molitor. Biochemistry 38, 11415–11424. Liou, Y.-C., Tocilj, A., Davies, P.L., Jia, Z., 2000. Mimicry of ice structure by surface hydroxyls and water of a β-helix antifreeze protein. Nature 406, 322–324. Olsen, T.M., Duman, J.G., 1997a. Maintenance of the supercooled state in overwintering Pyrochroid beetle larvae Dendroides canadensis: role of hemolymph ice nucleators and antifreeze proteins. Journal of Comparative Physiology B 167, 105–113. Olsen, T.M., Duman, J.G., 1997b. Maintenance of the supercooled state in the gut of overwintering Pyrochroid beetle larvae, Dendroides canadensis: role of gut ice nucleators and antifreeze proteins. Journal of Comparative Physiology B 167, 114–122. Olsen, T.M., Sass, S.J., Li, N., Duman, J.G., 1998. Factors contributing to seasonal increases in inoculative freezing resistance in overwintering fire-colored beetle larvae Dendroides canadensis (Pyrochroidae). Journal of Experimental Biology 201, 1585–1594.
111
Raymond, J.A., DeVries, A.L., 1977. Adsorption inhibition as a mechanism of freezing resistance in polar fishes. Proceedings of the National Academy of Sciences USA 86, 881–885. Raymond, J.A., Wilson, P.W., DeVries, A.L., 1989. Inhibition of growth on nonbasal planes in ice by fish antifreeze. Proceedings of the National Academy of Sciences USA 86, 881–885. Sicheri, F., Yang, D.S.C., 1995. Ice-binding structure and mechanism of an antifreeze protein from winter flounder. Nature 375, 427–431. Sjursen, H., So¨ mme, L., 2000. Seasonal changes in tolerance to cold and desiccation in Phauloppia sp. (Acari, Oribatida) from Finse, Norway. Journal of Insect Physiology 46, 1387–1396. Smallwood, M., Worral, D., Byass, L., Elias, L., Ashford, D., Doucet, C.J., 1999. Isolation and characterization of a novel antifreeze protein from carrot (Daucus carota). Biochemical Journal 340, 385– 391. Sun, X., Griffith, M., Pasternak, J.J., Glick, B.R., 1995. Low temperature growth, freezing survival, and production of antifreeze protein by the plant growth promoting rhizobacterium Pseudomonas putida GR 12-2. Canadian Journal of Microbiology 41, 776–784. Tursman, D., Duman, J.G., Knight, C.A., 1994. Freeze tolerance adaptations in the centipede Lithobius forficatus. Journal of Experimental Zoology 268, 347–353. Tursman, D., Duman, J.G., 1995. Cryoprotective effects of thermal hysteresis protein on survivorship of frozen gut cells from the freeze tolerant centipede Lithobius forficatus. Journal of Experimental Zoology 272, 249–257. Urrutia, M.E., Duman, J.G., Knight, C.A., 1992. Plant thermal hysteresis proteins. Biochimica et Biophysica Acta 1121, 199–206. Wilson, P.W., 1994. A model for thermal hysteresis utilizing the anisotropic interfacial energy of ice crystals. Cryobiology 31, 406–412. Wu, D.W., Duman, J.G., 1991. Activation of antifreeze proteins from the beetle Dendroides canadensis. Journal of Comparative Physiology 161, 279–283. Wu, D.W., Duman, J.D., Xu, L., 1991. Enhancement of insect antifreeze protein activity by antibodies. Biochimica et Biophysica Acta 1076, 416–420. Zachariassen, K.E., Husby, J.A., 1982. Antifreeze effect of thermal hysteresis agents protects highly supercooled insects. Nature 298, 865–867.