Many coregonine species have declined drastically across the Northern Hemisphere, including populations of Coregonus maraena (whitefish) in the Baltic Sea, and the mechanisms leading to these declines are not well investigated. An abrupt population crash occurred in the 1990s, coinciding with heavy declines in salmonid recruitment, also known as thiamine deficiency syndrome. Offspring with thiamine deficiency have a high mortality, posing significant negative impact on populations. Here, we aim to determine if whitefish, like other salmonids in the Baltic Sea, are affected by thiamine deficiency. Anadromous whitefish were therefore sampled during spawning in rivers of Southeastern Sweden, and we compared tissue concentrations and thiamine-dependent enzyme latencies to published thresholds. Further, we tested whether the variation in thiamine concentrations among individuals could be explained by physiological and morphological traits. Results showed that latency of thiamine-dependent enzymes, along with egg thiamine concentrations, suggests no evident thiamine deficiency. Concentrations were generally higher in the liver compared to muscle tissues. While females had lower liver thiamine concentrations compared to males, the opposite was found for muscle tissues, suggesting sex-specific patterns of allocation of the vitamin. Concentrations in eggs were positively related to the condition of the females and, similar to muscle and liver tissues, tended to be negatively related to standardized gill raker length. The latter is often used as a proxy for characterizing the feeding niche of coregonines. As has been observed in a number of other organisms (e.g., fish and molluscs), there was a reduction in thiamine concentration with length. Hence, the populations studied here showed no evidence of exhibiting thiamine deficiency. The variation in thiamine concentrations could largely be attributed to intrinsic physiological traits as well as traits associated with coregonine feeding niche.

Study system and sampling

Adult whitefish were sampled from three anadromous populations during spawning (November to December, 2020) in the rivers Lyckebyån, Skräbeån, and Virån along the Swedish south-east coast (Fig. 1A). In Virån (Kalmar County, n = 39), whitefish were caught using gill nets and cast nets. In Lyckebyån (Blekinge County, n = 20), dip nets were applied, whereas electrofishing and gill nets were used in Skräbeån (Skåne County, n = 20). Additional female whitefish were caught in Virån (n = 13; 2020) and Skräbeån (n = 8; 2022) and stripped for egg samples only before being released back into the wild.

Each specimen collected for determining tissue thiamine concentrations was euthanized by percussion stunning and severing the gill arch to quickly bleed out. Afterwards, each specimen was sexed, weighed, and its total length was noted. We then eviscerated the fish and weighed the liver and the gastrointestinal tract separately. To investigate their thiamine status, we sampled a piece of dorsal muscle, liver, and eggs. As females were ovulating at the time of sampling, their gonads were (partly) emptied. Hence, we could not sample eggs from all females, and gonad weight could not be reliably determined. All tissue samples were transported on dry ice and stored at -80 °C. The eviscerated whitefish were stored at -20 °C for further processing (see section “Gill raker count and length” below). The sampling was conducted under ethical permit (5.2.18-482/14).

Thiamine analysis

Thiamine was analyzed in muscle, liver, and eggs according to (Brown et al. 1998) with slight modifications. For a detailed description of the process, see Todisco et al. (2024). While liver and egg samples were analyzed as described in Todisco et al. (2024), muscle tissue (~ 1 g) was analyzed in half the stated volume of 2% and 10% trichloroacetic acid, as thiamine concentrations in muscle tissue were comparatively low. In short, a subsample of tissue was homogenized and boiled in diluted trichloroacetic acid. After centrifugation, the supernatant was washed with a mixture of ethyl acetate and hexane. Lastly, we added the dye K3Fe(CN)6 and filtered the mixture before it was injected into a Hitachi Chromaster HPLC system to measure fluorescence. We quantified three thiamine vitamers: free thiamine (TF), thiamine monophosphate (TMP), and thiamine diphosphate (TDP). Their concentrations were normalized per g wet weight and summed to total thiamine concentration (Ttot). Vitamer ratios were studied in muscle and liver samples, but since whitefish eggs partly thawed during sample preparation, shifts in vitamer ratios may have occurred. Hence, we refrained from studying vitamer ratios in eggs.

Transketolase activity

The enzymatic functionality of transketolase is dependent on TDP binding to it. In the matrix of thiamine-deficient organisms, transketolase lacking TDP may be present, which renders these enzymes inactive. Measurements for the presence of such unsaturated transketolase, i.e., latency, can be used as an indicator for the thiamine status of an organism. Commonly, latencies above 20% are categorized as thiamine-deficient (Jones et al. 2021).

We measured liver transketolase activity and latency using the BCA Protein Assay Kit (ab102536, Abcam) and Transketolase Activity Assay Kit (ab273310, Abcam) as described in Hauber et al. (under review), including modifications from (Jones et al. 2021). In short, liver tissues were homogenized in Tris Buffer, centrifuged and filtered. Samples were diluted to reach protein concentrations between 0.2-0.4 µg/µl. Following the manufacturer's protocol, assays for measurement of the basal transketolase activity were prepared. We added assays infused with TDP to measure the saturated transketolase activity. Standards, substrate control, sample background controls, and positive control were prepared following the manufacturer’s protocol. Using the microplate reader (FLUOstar Omega, BMG Labtech) we measured fluorescence every minute for 60 min. Following the manufacturer’s protocol, we calculated an average basal and stimulated transketolase activity for each specimen. The latency was calculated by dividing the basal activity by stimulated activity, subtracting it from 1 and multiplying it by 100.

Three of the samples were run on a separate plate and show much lower, only a quarter as high, transketolase activities. We believe this difference between plates is not of biological origin but caused by a misstep during manufacturing or protocol preparation. Hence, we excluded these three samples from the statistical analysis.

Egg coloration, weight, and water content

We measured egg parameters to assess whether thiamine concentrations relate to egg quality, to get a better understanding of the variation in egg thiamine concentrations.

Egg pigmentation, particularly the presence of carotenoids, was measured as an additional indicator of egg quality. Around 1 g of thawed eggs was spread out in an aluminium cup and photographed from above in a standardized setup under consistent overhead lighting in a windowless room (Canon EOS 700D; ISO 200, shutter speed 1/25 s). In Photoshop (version 21.1.0 © 2020 Adobe), the area covered by eggs in each picture was selected, and we measured an average value in the CIE (Commission Internationale de l’Eclairage) L*a*b* color space. This standardized, uniform, and device-independent color space separates luminescence (L*) from two measures of color intensity (a* and b*). Both channels a* and b* can be used to indirectly measure the presence of pigments (Svensson et al. 2005, Brüsin et al. 2016). The a* channel correlates with red carotenoids such as astaxanthin, the b* channel correlates with yellow carotenoids such as lutein (Hatlen et al. 1998, Humphries et al. 2004).

After the eggs were photographed, they were dried at 60 °C for 48 h and weighed again to calculate their water content. Lastly, the number of eggs was counted to calculate the weight per egg.

Gill raker count and length

Even within the same environment, whitefish can display variation in phenotype and niche utilization, resulting in divergent ecotypes. These ecotypes differ in functional morphological traits that correlate with habitat use and foraging strategies. In particular, the count and size of protrusions from the gill arches, i.e., gill rakers, may indicate the usage of different trophic niches (Ostbye et al. 2005, Kahilainen and Ostbye 2006).

To study the gill raker count and length, we thawed the eviscerated fish and removed the first left gill arch. We took standardized pictures (Canon EOS 700D) of the gill arch and counted the number of gill rakers (Fig. 1C). Lastly, we measured the length of the first (from top) gill raker on the lower gill arch (Fig. 1C) using Fiji (an ImageJ distribution, version 1.53c, Schindelin et al. 2012). We could not determine the number of gill rakers for one individual from Lyckebyån, as the gill arch had been damaged.

Statistical analysis

To test the effect of physiological as well as morphological traits on thiamine concentrations of somatic tissues, i.e., muscle and liver, we fitted a linear model with liver and muscle thiamine concentrations as the response variable. Tissue type, population, sex, length, condition, standardized gill raker length, and gill raker count were included as fixed effects, assuming a Gaussian error distribution. Interaction terms were added for tissue type with population and sex. Lastly, we included a unique identifier for each fish as a random intercept using the lme4 package (Bates et al. 2015). We calculated the condition, used as a proxy for health, by fitting total weight against length, both log-transformed, and extracting the residuals. We standardized gill raker length in the same way.

To investigate whether maternal effects or egg quality relate to thiamine concentrations in reproductive tissues, i.e., eggs, we fitted two linear models with egg thiamine concentrations as the response variable. We ran two models, since some egg samples were from fish that were released back into the wild, and therefore we were lacking maternal data. Maternal effects on egg thiamine concentrations were tested by including length, condition, standardized gill raker length, and muscle and liver thiamine concentrations as fixed effects. To test whether egg quality relates to egg thiamine concentrations, we included channel a*, channel b*, egg water content, and weight per egg as fixed effects.

Lastly, transketolase activity and latency were used as response variables in two separate models that included length, condition, sampling location, and liver thiamine concentration as fixed effects. Length was scaled in all models. We evaluated model fit using diagnostic plots of residual distribution with the DHARMa package (Hartig 2024). P-values were computed based on robust covariance matrix estimation, i.e., Wald tests. All statistical analyses were performed using R (version 4.4.2, R Core Team 2024).

References

Bates, D., M. Mächler, B. Bolker, and S. Walker. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67:1-48.

Brown, S. B., D. C. Honeyfield, and L. Vandenbyllaardt. 1998. Thiamine Analysis in Fish Tissues. American Fisheries Society Symposium 21:73-81.

Brüsin, M., P. A. Svensson, and S. Hylander. 2016. Individual changes in zooplankton pigmentation in relation to ultraviolet radiation and predator cues. Limnology and Oceanography 61:1337-1344.

Hartig, F. 2024. DHARMa: Residual Diagnostics for Hierarchical (Multi-level / Mixed) Regression Models.

Hatlen, B., M. Jobling, and B. Bjerkeng. 1998. Relationships between carotenoid concentration and colour of fillets of Arctic char, Salvelinus alpinus (L.), fed astaxanthin. Aquaculture Research 29:191-202.

Hauber, M. M., V. Todisco, O. Nordahl, P. Tibblin, E. Fridolfsson, E. Kärvegård, and S. Hylander. under review. Thiamine Allocation and Deficiency Status throughout the Life Cycle of Cod.

Humphries, J. M., R. D. Graham, and D. J. Mares. 2004. Application of reflectance colour measurement to the estimation of carotene and lutein content in wheat and triticale. Journal of Cereal Science 40:151-159.

Jones, K. S., D. A. Parkington, L. J. Cox, and A. Koulman. 2021. Erythrocyte transketolase activity coefficient (ETKAC) assay protocol for the assessment of thiamine status. Annals of the New York Academy of Sciences 1498:77-84.

Kahilainen, K., and K. Ostbye. 2006. Morphological differentiation and resource polymorphism in three sympatric whitefish Coregonus lavaretus (L.) forms in a subarctic lake. Journal of Fish Biology 68:63-79.

Ostbye, K., T. F. Naesje, L. Bernatchez, O. T. Sandlund, and K. Hindar. 2005. Morphological divergence and origin of sympatric populations of European whitefish (Coregonus lavaretus L.) in Lake Femund, Norway. Journal of Evolutionary Biology 18:683-702.

R Core Team. 2024. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

Schindelin, J., I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, J. Y. Tinevez, D. J. White, V. Hartenstein, K. Eliceiri, P. Tomancak, and A. Cardona. 2012. Fiji: an open-source platform for biological-image analysis. Nature Methods 9:676-682.

Svensson, P. A., E. Forsgren, T. Amundsen, and H. N. Sköld. 2005. Chromatic interaction between egg pigmentation and skin chromatophores in the nuptial coloration of female two-spotted gobies. Journal of Experimental Biology 208:4391-4397.

Todisco, V., E. Fridolfsson, C. Axén, E. Dahlgren, M. J. Ejsmond, M. M. Hauber, K. Hindar, P. Tibblin, M. Zöttl, L. Söderberg, and S. Hylander. 2024. Thiamin dynamics during the adult life cycle of Atlantic salmon (Salmo salar). Journal of Fish Biology 104:807-824.