Machine learning highlights the importance of primary and secondary production in determining habitat for marine fish and macroinvertebrates

1. Species distribution models for marine organisms are increasingly used for a range of applications, including spatial planning, conservation, and fisheries management. These models have been constructed using a variety of mathematical forms and drawing on both physical and biological independent variables; however, what might be called first-generation models have mainly followed the form of linear models, or smoothing splines, informed by data collected in the context of fish surveys.
2. The performance of different classes of variables were tested in a series of species occurrence models built with machine learning methods, specifically evaluating the potential contribution of lower trophic level data. Random forest models were fitted based on the classification of the absence/presence for fish and macroinvertebrates surveyed on the US Northeast Continental Shelf.
3. The potential variables included physical, primary production, secondary production, and terrain variables. For accepted model fits, six variable importance measures were computed, which collectively showed that physical and secondary production variables make the greatest contribution across all models. In contrast, terrain variables made the least contribution to these models.
4. Multivariable analyses that account for all performance measures reinforce the role of water depth and temperature in defining species presence and absence; however, chlorophyll concentration and some specific zooplankton taxa, such as Metridia lucens and Paracalanus parvus, also make important contributions with strong seasonal variations.
5. Our results suggest that lower trophic level variables, if available, are valuable in the creation of species distribution models for marine organisms.

     

Porcine theca cells produce immunoreactive beta-endorphin and change steroidogenesis in response to opioid agonist.

In earlier in vitro experiments opioids affected steroidogenesis in porcine luteal and granulosa cells. The present studies were undertaken to examine the effects of FK 33-824 (opioid agonist) alone or in combination with LH, PRL or naloxone (NAL, opioid antagonist) on steroidogenesis in cultured porcine theca cells. Moreover, we have tested beta-endorphin-like immunoreactivity (beta-END-LI) concentrations in culture media under control conditions and following treatments of theca cells with LH, PRL, progesterone (P4), oestradiol (E2) or testosterone (T). FK 33-824 and NAL significantly increased P4 release by theca cells and inhibited stimulatory effect of LH on this steroid output. PRL-induced P4 secretion from the cells was blunted only by FK 33-824. Secretion of androstenedione (A4) and T was essentially elevated in the presence of FK 33-824 and this potentiation of both androgen release was completely abolished by PRL. NAL blocked stimulatory effect of the opioid agonist only in case of T. Secretion of oestradiol and oestrone was completely free from the influence of both the opioid agonist and antagonist. Pig theca cells were able to produce beta-END-LI but none of tested hormones (LH, PRL, P4, E2 and T alone or in combination) significantly affected this production. In conclusion, these data indicate that porcine theca cells may produce beta-END-LI and change their steroidogenesis in response to opioid peptides.     

Exocrine pancreatic secretion in pigs fed sow's milk and milk replacer, and its relationship to growth performance

The objective of this study was to quantify and compare the effects of sow's milk and 2 milk replacer diets (containing clotting or non-clotting protein sources) on exocrine pancreatic secretion, plasma cholecystokinin, and immunoreactive cationic trypsin in pigs. In addition, the relationship between exocrine pancreatic secretion and growth in milk-fed pigs was studied. In a changeover experiment, 9 chronically catheterized pigs of 6.6 +/- 0.19 kg of BW were studied for 3 wk. Pigs were assigned to each of 3 diets. Exocrine pancreatic secretion was measured from the third to the seventh day on each diet. The protein content and trypsin activity of the pancreatic juice were measured. Blood samples were taken at 10 min before and after milk ingestion and were analyzed for cholecystokinin and immunoreactive cationic trypsin. Pancreatic protein and trypsin secretion did not differ between pigs fed sow's milk and those fed milk replacer, but the volume secreted was less for the pigs fed sow's milk (0.75 vs. 1.03 mL x kg(-1) x h(-1); P < 0.01). A postprandial response to milk intake was not observed. The 2 milk replacer diets did not affect exocrine pancreatic secretion differently. The average exocrine pancreatic secretion (volume, 0.94 mL x kg(-1) x h(-1); protein, 4.28 mg x kg(-1) x h(-1); trypsin, 1.65 U x kg(-1) x h(-1)) was intermediate between literature values for suckling and weaned pigs. Plasma cholecystokinin was elevated (approximately 18 pmol x L(-1)) and showed low correlations with the pancreatic secretion traits. Plasma immunoreactive cationic trypsin was not significantly related to any of the pancreatic secretion traits and should therefore not be used as an indicator for exocrine pancreatic function in milk-fed pigs. Exocrine pancreatic secretion varied substantially among individual pigs (protein, 0.22 to 13.98 mg x kg(-1) x h(-1)). Pancreatic protein and trypsin secretion showed a positive, nonlinear relationship with performance traits. It was concluded that neither specific sow's milk ingredients nor the protein source are responsible for a low pancreatic protein secretion in suckling pigs. Exocrine pancreatic secretion was positively correlated with ADG in pigs at an identical milk intake.     

Factors influencing fruit size in the strawberry (Fragaria ananassa Duch.)

Summary Fruit size is one of the most important characteristics of highly productive strawberry cultivars. The aim of the experiments was to establish the factors responsible for the expression of this trait. The total yield of large fruits correlates closely with the total yield of all fruits and depends mainly on the mean fruit weight of all fruits. The position of fruits on the inflorescence influences the decline of fruit size to a larger extent in small-fruited clones than in the large-fruited ones. The size of the fruit is controlled by the dimension of the receptacle and number of achenes. The stimulating effects of achenes are quite different in various genotypes and the fruit weight per achene declines with the inferior blossom position. The large-fruited clones have bigger leaves, a larger photosynthetic area, and thicker petioles and flower stalks. Their cells are larger, which is common for the giant genotypes. The results suggest that there exists a possibility for indirect selection of fruit size on the basis of some additional parameters, which can be useful particularily in the years of unsuitable weather conditions.     

Relationship between somaclonal variation and type of culture in cucumber

Highly inbred B line of cucumber was used to compare the effect of four types of in vitro culture on somaclonal variation. The plants were regenerated from the following types of culture: twelve- and eighteen-month-old liquid culture of meristematic clumps (LMC₁₂₍₁₈₎), ten-month-old embryogenic cytokinin-dependent suspension (CDS), eighteen-month-old embryogenic cytokinin-dependent suspension in medium with modified NH⁺ ₄/NO₃ ^- ratio (CDS 1.7), twelve-month-old embryogenic auxin-dependent suspension (ADS), thirty six-month-old embryogenic auxin-dependent suspension in medium with modified NH⁺ ₄/NO₃ ^- ratio (ADS 1.7) and recurrent leaf callus regeneration (RLC) – repeated 5 times. The differences in the incidence of the following properties were observed: the ploidy of R₀ plants, the segregation of new morphological traits in R₁ and the germination ability of R₁ seeds. R₁ families with the segregation of new phenotypes were most numerous in CDS (62.5%) and LMC₁₈ (57.9%), next in CDS1.7 (35.7%), while the smallest number was found in LMC₁₂ (11.1%) and RLC (3.4%).Tetraploid and mixoploid plants occurred in ADS1.7 and ADS (100%) whereas CDS and RLC were observed to contain only tetraploids, respectively 33.3% and 55.2%. There were no changes of ploidy after LMC₁₂, LMC₁₈ and CDS1.7. Among new phenotypes there were such that have not been described so far in cucumber: ginkgolike leaf (gll), yellow-green chlorophyll mutants (y-gc), serrate margin of corolla in male and female flowers (smc). This revised version was published online in August 2006 with corrections to the Cover Date.     

Influence of selenium on oxidoreductive enzymes activity in soil and in plants

The aim of this greenhouse experiment was the assessment of the influence of H₂SeO₃ at soil concentrations of 0.05, 0.15 and 0.45 mmol kg⁻¹, on the activity of selected oxidoreductive enzymes in wheat (Triticum aestivum). The wheat plants were grown in 2 dm³ pots filled with dust-silt black soil of pH 7.7. Applied H₂SeO₃ caused activation of plant nitrate reductase at all concentrations, but activation of plant polyphenol oxidase at only two lower concentrations. The highest concentration caused inhibition of polyphenol oxidase and peroxidase. Plant catalase activity decreased under the influence of 0.15 and 0.45 mmol kg⁻¹ concentration. After the final analysis Se was quantified in plants and soil. The amounts in plants were: control (unamended soil) 1.95 mg kg⁻¹; I dose (0.05 mmol kg⁻¹) 18.27 mg kg⁻¹; II dose (0.15 mmol kg⁻¹) 33.20 mg kg⁻¹ and III dose (0.45 mmol kg⁻¹) 38.37 mg kg⁻¹, in soil: 0.265 mg kg⁻¹; 3.61 mg kg⁻¹; 10.53 mg kg⁻¹; 30.53 mg kg⁻¹; respectively. Simultaneously, a laboratory experiment was performed, where the activity of soil catalase and peroxidase were tested after 1, 3, 7, 14, 28, 56, and 112 days after Se treatment. Peroxidase activity in soil decreased with increasing Se content, over the whole experiment. The lowest dose of Se caused activation a significant 10% increase in catalase activity, but the influence of others doses was unclear.     

Mosses as bioindicators of heavy metal pollution in polish national parks

The concentrations of heavy metals (Cd, Co, Cr, Ni, Cu, Ph, Zn, Mn, Fe) and other elements (Mg, Na, K, Ca) were determined in the samples of Pleurozium schreberi and Hylocomium splendens from 12 Polish national parks. The significant differences in the concentrations of all heavy metals between particular parks were found. The lowest concentrations of these metals were recorded in the mosses from the national parks in northern Poland, the highest in southern Poland. Significant differences in the heavy metals contents of the mosses between central and peripheral park areas, between green and brown parts of mosses, as well as between moss species were also found.     

Fatty acid composition of various ectomycorrhizal fungi and ectomycorrhizas of Norway spruce

Whole cell fatty acid (WCFA) compositions of three different structures of ectomycorrhizal (ECM) fungi: sporocarps, pure culture mycelia and ectomycorrhizas were analysed to evaluate the potential use of fatty acid profiles as biomarkers for ECM fungi and ectomycorrhiza-associated bacteria. Sporocarps of Amanita muscaria, Amanita rubescens, Lactarius rufus, Lactarius thejogalus, Leccinum scabrum, Paxillus involutus, Russula foetens, Russula rosea, Russula vesca, Suillus grevillei, Tylopilus felleus, Xerocomus badius, Xerocomus subtomentosus, pure cultures of A. muscaria, P. involutus, X. badius, X. subtomentosus, Suillus bovinus Suillus luteus and seven ectomycorrhizal morphotypes of Norway spruce were examined. Our results revealed species-specific composition of fatty acids of fungal sporocarps and pure culture mycelia. Ectomycorrhizal morphotypes distinguished and identified by morphological and molecular methods (PCR-RLFP and sequencing) created specific fatty acid profiles. The dominating fatty acids in pure cultures and sporocarps were 18:2ω6,9, 18:1ω9 and 16:0, whereas ectomycorrhizas also contained plant and bacterial specific fatty acids. Especially, fatty acids specific to Gram-positive bacteria 15:0 anteiso and 17:0 anteiso were present in relatively high amounts and suggested that these bacteria are dominating in the examined Norway spruce mycorrhizosphere. In conclusion, our results show that fatty acid based methods can be useful in studies of ectomycorrhizal fungi, both as a quick method for differentiation of fungal species and also in studies of mycorrhiza-associated microorganisms in the field.     

Conservation-oriented forestry and early successional saproxylic beetles: Responses of functional groups to manipulated dead wood substrates

New conservation-oriented forestry aims to maintain intact populations of forest organisms by improving the conservation value of managed forests and providing protected areas. We tested the conservation value of treatments of dead wood for assemblages of early successional saproxylic beetles. In nine areas in northern Sweden, we selected one clear-cut, one mature managed forest and one reserve. In 2001-2002, we placed three blocks of spruce logs, each containing control, burned and shaded logs and a high stump (“snag”) at each site. Saproxylic beetles emerging from the dead wood were collected using emergence traps and beetles flying close to it were collected using flight-intercept traps. After one year of exposure, assemblage composition was examined, with respect to nutritionally-defined functional groups, red-listed species and fire-favoured species. Experimental snags were most complementary to control logs, supporting different assemblages of cambium consumers and fungivores and supporting more red-listed individuals. Burned logs supported depauperate assemblages, particularly with respect to cambium consumers, while shading of logs affected assemblages of fungivores, but only on clear-cuts. Despite containing less dead wood, managed forests provided valuable habitat, supporting similar assemblages of saproxylic beetles to reserves. Most functional groups were less abundant on clear-cuts than in older forests, but fire-favoured species were more common on clear-cuts, suggesting that clear-cuts may support assemblages of species associated with natural disturbances, if suitable substrates are available. Utilization of logs by saproxylic beetles changes over time, so long-term monitoring of our experimental logs will determine their lifetime conservation value.     

Near-infrared spectroscopy for analysis of chemical and microbiological properties of forest soil organic horizons in a heavy-metal-polluted area

In industrial areas, heavy metals may accumulate in forest soil organic horizons, affecting soil microorganisms and causing changes in the chemical composition of the accumulated organic matter. The objectives of this study were to test the ability of near-infrared spectroscopy (NIRS) to detect heavy metal effects on the chemical composition of forest soil O horizons and to test whether NIRS may be used to quantitatively determine total and exchangeable concentrations of Zn and Pb (Zn_t, Pb_t, Zn_ex, Pb_ex) and other chemical and microbial properties in forest soil O horizons polluted with heavy metals. The samples of O horizons (n = 79) were analyzed for organic C (C_org), total N and S (N_t, S_t), Zn_t, Pb_t, Zn_ex, Pb_ex, basal respiration (BR), microbial biomass (C_mic) and C_mic-to-C_org ratio. Spectra of the samples were recorded in the Vis-NIR range (400–2,500 nm). To detect heavy-metal-induced changes in the chemical composition of O horizons principal components (PC1–PC7) based on the spectral data were regressed against Zn_t + Pb_t values. A modified partial least squares method was used to develop calibration models for prediction of various chemical and microbial properties of the samples from their spectra. Regression analysis revealed a significant relationship between PC3 and PC5 (r = −0.27 and −0.34, respectively) and Zn_t + Pb_t values, indicating an effect of heavy metal pollution on the spectral properties of the O horizons and thus on their chemical composition. For quantitative estimations, the best calibration model was obtained for C_org-to-N_t ratio (r = 0.98). The models for C_org, N_t, and microbial properties were satisfactory but less accurate. NIRS failed to accurately predict S_t, C_org-to-S_t, Zn_t, Pb_t, Zn_ex, and Pb_ex.