Thiamine (vitamin B₁) has major impacts on survival of embryonic and early life stages of fish. Low levels of thiamine, a water-soluble vitamin, in eggs have been responsible for early mortality syndrome (EMS) in salmonids from the lower Great Lakes (Marcquenski, 1996), Cayuga syndrome in Atlantic salmon Salmo salar from the Finger Lakes region (Fisher et al., 1995), and M74 (another EMS-like cause of mortality) in Baltic Sea salmon (Bengtsson et al., 1994). Fisher et al. (1996) concluded that the increase in salmonid alevins mortality from Cayuga Lake was directly linked to broodfish diets consisting only of alewife Alosa pseudoharengus and rainbow smelt Osmerus mordax. These two nonindigenous species, as dominant forage fishes in Cayuga Lake, contain thiaminase, an enzyme that destroys thiamine (Fisher et al., 1995, 1996). Similar conclusions were drawn by Fitzsimons and Brown (1998) with respect to lake trout Salvelinus namaycush in the Great Lakes. Alewife and rainbow smelt also have been reported to be the main prey of walleye in northern Lake Michigan (Madenjian et al., 2002) and the western basin of Lake Erie (Morrison et al., 2000), leading to the potential that walleye Stizostedion vitreum could suffer EMS.

Therefore, we hypothesize that thiamine deficiency in walleye also can occur. Alevins from Lake Michigan coho salmon Oncorhynchus kisutch with high and medium survival were found with similar levels of total thiamine, but fish with low survival had significantly lower levels (Honeyfield et al., 1998, Fig. 1). These same authors reported that eggs from fish with high and medium survival had higher levels of free thiamine and thiamine monophosphate than eggs from fry with low survival. In contrast, levels of thiamine pyrophosphate did not differ among the three groups (Fig. 1).

Hishida and Nakano (1954) were the first to demonstrate the importance of thiamine mobilization into a phosphated form during embryonic development of fish. These data support the hypothesis that low egg thiamine level is an important factor in EMS. Therefore, the better we understand the deposition process of thiamine into the egg, the better we will understand EMS ethiology. Egg concentrations of total thiamine strikingly differed between Lake Michigan coho salmon (1.5-6.3 nmol/g) and captive broodstock (29.8 nmol/g) fed a thiamine-replete diet (Honeyfield et al., 1998). Fitzimmons (1995) first reported that yolk sac alevins survival was improved when fish were treated with thiamine. Bylund and Lerche (1995) also reported that Atlantic salmon eggs or fry treated with thiamine solution (500 mg/l) significantly improved their survival. Although thiamine therapy offers an immediate treatment for EMS, it treats only the symptom, and not the cause of the problem. Therefore, determining the underlying cause of EMS would be a step toward understanding and addressing reproductive dysfunctions affecting the Great Lakes predatory fishes.

Clearly vitamin B₁ is critical to fish reproductive processes. The effect of nutrient availability on offspring viability in Great Lakes salmonid fish diets is well documented. However, few data exist regarding possible impacts of vitamin B₁ deficiency on gamete biochemical characteristics in walleye and the viability of their eggs or fertilizing ability of their sperm. Understanding of nutrient-regulated success in reproductive output in walleye is a prerequisite for further work on accurate prediction of population dynamics. If we are able to show correlation between concentration of thiamine and total number of viable eggs produced by the population, specific river stocks, in a particular year (age class), these data could be used to interpret variability in walleye recruitment.

4.1. Fish collection and sampling

Adult walleye will be collected at four different sites, including Maumee River, Sandusky River, Grand River, and Western-basin reefs of Lake Erie, during their spawning season (April-May). Fish will be captured by personnel from the Ohio Division of Wildlife, Department of Natural Resources, using electrofishing in streams or using gillnets in Lake Erie. Ten "running" (ready to ovulate) females and ten spermiating males at each site will be individually spawned by stripping gametes on site. Fish will be individually weighed and measured. Otoliths will be sampled for age determination. Samples of eggs from individual, completely ovulated females will be collected by stripping prior to fertilization and immediately frozen (-40°C; dry ice) for further biochemical analysis. Viable gametes will be stored in individual, flat-bottom containers at 5-7°C (eggs, 2-3 layers of eggs) or 0°C (sperm, 1-2 mm of sperm) during transport to the Columbus laboratory. On arrival to the laboratory, sperm concentrations will be estimated microscopically using a Double Neubauer Counting Chamber (Ciereszko and Dabrowski, 1993). Sperm will be centrifuged at 7,280 g for 10 min at 4°C. Seminal plasma and spermatozoa will be separated and frozen for further biochemical analysis.

4.2. Evaluation of gamete quality

To evaluate egg quality, eggs (2 g or ~ 400 eggs) will be fertilized in duplicate with the milt of five males (25,000 spermatozoa/egg) using the dry method previously described by Czesny and Dabrowski (1998). This sperm concentration has been experimentally estimated in preliminary studies (Fig. 3) to provide critical sperm density for quality evaluation (Rurangwa et al., 1998). Sperm will be diluted in Moore extender prior to fertilization based on prior estimation of individual male sperm density. To prevent egg-to-egg adhesion, eggs will be treated with a solution of tannic acid (Sigma, St. Louis, MO) at 400 mg/l for 4 min with continuous stirring. Fertilized eggs from each female will be incubated separately in a circular basket (2 baskets per female) with a net bottom in "California type" incubators. Dechlorinated city water at 15°C will be used in a recirculated system equipped with a UV-sterilizing system and 20% water exchange rate per day. The rate of survival will be assessed at the pigmented eyed-stage embryo and at hatching (McElman and Balon, 1979). At hatching, larvae will be fixed in glutaraldehyde and cacodylate fixative for further measurement and analysis (Dabrowski and Bardega, 1982). In order to standardize a hatching stage, baskets will be transferred into aerated tank with stagnant water and larvae will be collected within 2-3 hours. To evaluate sperm quality, eggs from 5 females will be combined and fertilized with sperm from individual males, incubated and eyeing and hatching success will be determined as described previously.

4.3. Biochemical analysis

Biochemical analysis of gametes will include analysis of vitamin B₁. Moreover, lipid and fatty acid analysis also will be performed for our project supported by the Ohio Department of Natural Resources.

4.3.1. Vitamin B₁ analysis

High performance chromatography (HPLC) and conventional thiochrome methods (Hennessy and Cerecedo, 1939) will be used to determine concentrations of vitamin B₁ and its derivatives in gametes (Brown et al., 1998). We are going to use HPLC (Beckman Instruments, Inc., Palo Alto, CA) and UV-amperometric detectors (Bioanalytical Systems, West Lafayette, IN) routinely used in our laboratory (Lee and Dabrowski, 2003).

4.3.2. Lipid and fatty acid analysis

Lipid and fatty acid analyses will be performed as previously in our laboratory (Czesny and Dabrowski, 1998). A sample of eggs (5 g) from individual females will be taken before fertilization and immediately frozen in liquid nitrogen, then stored in a biofreezer (-85°C) prior to biochemical analysis. Total lipids will be extracted from eggs after homogenization in chloroform-methanol according to the procedure of Folch et al. (1957). Total lipid extracts will be separated into polar (phospholipids) and neutral (mostly triglycerides) fractions using silica sep-pak cartridges (Waters, Division of Millipore Corp., Milford, MS) and determined gravimetrically. Fatty acid methyl ester mixtures will be prepared from both fractions of lipids according to the method described by Metcalfe and Schmitz (1961). The gas chromatogaph analyses of the fatty acid methyl esters will be performed on Varian 3300 Gas Chromatograph (Varian Chromatography Systems, Walnut Creek, CA) using internal standard (C19:0) as detailed by Czesny and Dabrowski (1998).

4.4. Statistical analysis

All data will be expressed as mean ± standard deviation or standard error. Statistical analyses will be performed using the Statistical Analysis System (SAS Institute, Inc., Cary, NC). Normality and homogeneity of variance will be verified prior to any statistical analysis. Percentage data will be arc sin transformed prior to statistical analysis. Significant differences between sites will be determined using a one-way analysis of variance (ANOVA) and subsequent comparison of means by Fisher least significant analysis. When assumptions of normality or equal variance fail to be met, nonparametric analysis (Kruskal-Wallis test) will be performed. Stepwise regression will be applied to all parameters to reveal significant interactions. Difference will be accepted as statistically significant when P<0.05.