Our project on the Atlantic cod, Gadus morhua, received funding in the Fall of 1996 and I am reporting the progress made by the end of a one year no cost extension of the project.

Progress Outline:

1. The Problem.
2. Objectives.
3. Data and samples from DEII 94-02, AL95-02, 96IsS.
4. Anti-Lv Serum Production and Characterization.
5. Participation in  NOAA Spring and Fall Bottom Survey Cruise.
6. Slime as a source of information on cod maturation.
7. Use of NMF facility  in Estradiol Induction of  Vitellogenin in captive cod.
8. Analysis of serum and slime sampled fish from AL98-11, AL99-03 and GL99-01.
9. Application of QIEP technology to DEII 94-02, AL95-02, 96IsS.
10. Conclusions.
11. Products.
12. Personnel Trained.
13. Plans for the Future.

1. The Problem.   During the fall bottom surveys of the Georges Bank and Gulf of Maine cod stocks, in particular, NMF staff have found it difficult to distinguish an immature female cod from a newly developing female.  Currently the determination is made by visual inspection of the ovary which is small and translucent in an immature female but which slowly starts developing venation and cloudy opaqueness and perhaps a yellowish orange coloration as maturation begins.   This difficult subjective change may be affected by local lighting and other extrinsic factors and makes it uncertain if a 50% maturation age can be accurately estimated and used as a gauge of whether cod stocks are changing in their age of maturation.
2. Objectives. It was proposed to develop an immunologic assay for Atlantic cod vitellogenin (Vg) and the derived lipovitellin (Lv) in order to establish an accurate method for aiding the determination of maturation state of the female.  It was supposed that a developing female would be synthesizing Vg and a quantitative assay would allow maturation state to be established.  To this end, the serum and mucus Vg titer of individual females will be measured and correlated with the maternal age, size and a visual as well as histological measure of the state of maturity of her ovary.  Particular emphasis will be placed on studying where the Georges Bank stocks fit in the 35-65 cm 'normal' age/size of females undergoing ovarian maturation among different cod stocks (as discussed in Morrison, 1987).
3. Data and samples from DEII 94-02, AL95-02, 96IsS.  So far as possible it was hoped to be able to provide methodology, analysis and insight into the extensive set of samples that had been collected in a series of three NMF cruises which predated the funding of this project.  The samples of serum and histological slides prepared by NMF from those three survey trips were transferred to the projects control.  Some of the serum samples were small and stored in large vials which led to freezing and thawing damage during storage at NMF.  Some of the serum had substantial hemolysis of red blood cells aside from being small in volume.  However many of the samples were of sufficient size that such damage was minimal.  In addition, other valuable experimental samples from the DEII 94-02 cruise were obtained including serum from animals injected with the protease inhibitor aprotinin.  These old serum samples were stored at UMass Amherst until a satisfactory methodology could be developed for assay of VG titer such that they could be analyzed and the Vg titer added to the extant maturation data which variably included weight, length, ovarian weight, liver weight, gutted weight and ovarian histology.
4. Anti-Lv Serum Production and Characterization.

After a preliminary survey of the sera (Quarterly Report #1) it was not clear that we could identify and purify the Vg from any of the extant samples of female serum.  We therefore sought to obtain fresh ovarian and egg material as a starting point for purification of lipovitellin, Lv, the egg storage form of Vg.   We obtained ovaries from commercial fishermen sailing out of Chatham MA.  Purification of Lv from these ovaries provided an antigen with which we made our first antiserum.  Initially we hoped to develop an ELISA test using this antiserum.  The antiserum proved to be specific for the purified cod Lv showing no reactivity to control male serum and immature female serum in Quantitative Immunoelectrophoresis (QIEP).  Using this antiserum in an ELISA test we were gratified that it was highly reactive with developing female serum and slime from developing females.  However, we were dismayed that there was a substantial background reaction with this antiserum to slime from all cod including males.  In an effort to solve this analytical problem we produced a second antiserum from Lv purified from eggs derived from a ripe female collected by my technician John Bohannon on an Albatross IV Spring Survey cruise (Quarterly Report #6).  The antiserum from that immunization was found to also react with components of slime from both males and females giving a false positive test.
We have thus rejected the use of an ELISA assay for cod LV and VG because of a background reaction to slime components.  QIEP, as reported before, remains specific for females and Vg and LV specifically.  Because undiluted cod slime comes with a high concentration of salt we can not use it directly in QIEP.  We developed a way of removing the salt while not diluting the slime proteins.  This technique thus allowed us to proceed in our tests of slime and serum for Vg content.
 
5. Slime as a source of information on cod maturation.  The proposal to use slime as a sampling tool to measure if a cod female is vitellogenic was based on the findings of Jennifer Specker on Striped Bass.  If we could sample slime rather than serum in cod, we would save a great deal of effort and have a relatively non-invasive technique for monitoring cod maturation.  Unfortunately, this approach turned out to be a blind alley.
6. Participation in  NOAA Spring and Fall Bottom Survey Cruise.

In the spring of 1998 my lab participated for the first time in a NMF Bottom Survey.  We needed to obtain fresh samples of cod eggs, slime, and blood serum from animals we could verify were maturing.  This participation in the survey allowed us to collect slime samples and some initial serum samples.  Initially, we sampled primarily slime and obtained a few serum samples, allowing the blood to clot and separating the serum by decantation.  This scheme proved inadequate.  By participating in the sampling myself, it became evident that the slime sampling process was fraught with the likelihood of slime contamination with cod blood.  The trauma of the 30 minute trawls created blood specks over the body of most cod examined.  Attempts to obtain slime from under the gill plate did not improve the likelihood of slime uncontaminated with blood.
Serum sampling, on the other hand, has become more efficient. In the earlier cod sampling cruises, cod were put into live tanks and sampling was done more leisurely. We did not have that option while we were fitting into the normal bottom survey routine in the 1998 Spring and Fall surveys and the 1999 Spring survey. However, improvements in sampling methodology worked out by Joe Zydlewsky, myself and Jay Burnett, resulted in our ability to sample as many as 30 animals from a single tow and obtain serum and slime samples from them all.  Briefly, sampling materials were prepared, pre numbered with expected sample numbers. This included, bandoliers of numbered syringes for serum bleeding and vials to receive slime samples plus boxes of pre numbered vials of Bouins fixative for ovarian slices.   The serum was allowed to clot before being centrifuged on a microfuge.  The separated serum was decanted from the cells into a fresh vial and frozen.
7. Use of NMF facility  in Estradiol Induction of  Vitellogenin in captive cod.  Our skepticism over the ability to use slime from trawled cod to measure Vg titer led us to design an experiment to test if cod females indeed secreted Vg into their serum.  Three of five original live cod were able to be maintained at the NMF aquarium facility at Woods Hole and were used in an experiment in which injected estrogen was used to induce vitellogenin production.  We sampled fish slime for Vg over long (weekly) and short term (daily) schedules after estradiol injection to see how responsive our slime QIEP assay will be for detecting slime-Vg changes in individuals.  The three cod were sacrificed at the end of the experiments and serum Vg determined.  Despite the fact that high titers of Vg were found in the cod serum at the end of the experiment, no VG was detected using our QIEP assay secreted in the slime.   This has several possible interpretations: (1) the measured Vg in cod slime is derived from trauma induced secretion of plasma components into slime only in trawl stressed individuals.  (2) Vg secretion into slime is a natural phenomenon but controlled by other hormones or conditions in the wild populations of fish which were not operant in our captive fish during the summer.
   A second captive cod experiment during the summer of 1999 was aimed at measuring how quickly cod respond to an estradiol injection. This experiment became possible because we had developed the technique of tail bleeding which was minimally traumatic to the cod and allowed daily bleeding.  After a single injection of 10 ug of estradiol, the single cod began responding after 4 days with measurable Vg in its serum.  On autopsy at the end of the experiment the cod was found to be a resting female.
8. Analysis of serum and slime sampled fish.  Slime from Bottom Survey cod were analyzed by QIEP and many cod females were shown to have Vg in their slime.  While encouraging, we went on to measure slime from cod caught by a commercial fishing boat which used long lines and brought up their fish with minimal trauma and the fish were maintained in a live tank for several days prior to our sampling; no observable blood specks were observed in their slime.  These slime samples did not show a positive test for Vg in our QIEP assay despite the fact that serum samples did contain measurable Vg.  This commercial catch of cod was done in May when almost all females were observed to be spent and serum Vg was at its lowest on average for the aggregate number of cod sampled.  We have, nonetheless, abandoned our pursuit of the usefulness of slime to monitor Vg production due to the uncertainties of relative trauma to individual fish and the additional uncertainties of interpretation of the measured titers.  Slime is clearly one source of blood from trawled cod and might be developed as a method to collect small blood samples if the mixture with slime components were not found to be a problem.

Our attention turned to using the serum titer of Vg as our probe for cod maturation.  Using the samples we had obtained from AL 98-11, AL 99-03 and GL 99-01, we have established a protocol for using our antiserum to aid in establishing the maturation state of cod females, Fig 1.   Immature females have no Vg in their serum. Developing females can have no Vg or very high Vg titer in their serum. Resting and Spent females have no Vg to medium levels of Vg in their serum. Surprisingly there were 4 of 210 males sampled that also had Vg in their serum. The quantities of Vg as viewed in Fig. 1 also have a seasonal variation. The resting females from our GL 99-01 survey done in May showed very low or zero Vg. Females from the April AL 99-03 survey that were classified as resting had larger maximums of Vg titer; which suggests that when cod females initially enter the resting phase they still may have Vg measurable in their serum but as time passes it is cleared.  This later observation and interpretation would have major implications for the Fall Bottom Surveys data interpretation; it would suggest that resting female Vg titers drop to zero during the summer and when a higher titer is measured in the fall, it represents Vg newly synthesised in the current season.
9. Data and samples from DEII 94-02, AL95-02, 96IsS.

Now that we have an approach that yields rational results from serum samples, we are in the process of applying the developed protocols to the frozen serum samples from the NMF cruises taken during 1994 to 1996.  These serum samples were collected on cruises that were designed more specifically for understanding cod maturation and have substantial ancillary data associated with them which will make the analyses more meaningful when completed.
10. Conclusions.

We have developed an approach to further define the maturation process in cod females by examining the Vg content of serum.  So far, the technique is not definitive in differentiating immature from developing females, the major practical objective of this project.  All ovaries labeled 'I' are indeed immature but some ovaries labeled 'D' may not be developing when judged with our anti-Vg-serum.  This may be because there is a stage of development of the ovary which precedes vitellogenesis. This may result in a sizable percentage of developing females being classified as immature by the antiserum, Table I. At the moment, we do not know whether the cutter determinations of D (developing) were false positives and those individual ovaries were I (immature) or whether a pre-vitellogenic stage of developing ovaries has been defined.  In addition, the basis for some resting (T) and spent (S) females having low to medium Vg titer is not understood.  Further analysis of the histology is being carried out to elucidate that issue.  An addendum to this final report will be submitted when we have sorted out the histology of the ovary samples which may explain the actual state of maturation of the cod, and which may create sub-categories of maturation such as 'pre-vitellogenic developing' ovaries.
11. Products.
    • A protocol for purification of cod lipovitellin.
    • Two antisera which are specific for cod lipovitellin.
    • A protocol for sampling cod serum on bottom surveys.
    • Samples of cod serum from AL-98-11, AL-99-03, and GL-99-01.
    • Samples of cod ovary fixed in Bouins from AL-99-03, and GL-99-01.
    • Preliminary measures of Vg in the serum of DEII 94-02, AL95-02, and 96IsS.
    • A WWW site with the accumulated data collected during this project available for download at URL:
      http://marlin.bio.umass.edu/biology/kunkel/fish/cod/
12. Personnel Trained:

Undergraduates: Corrie Perlroth, Robert Alperin-Lee, Mike Pelak, Ray Moniz, Rahul Sharma.
Graduate students: Ruth Hartling, John Bohannon, Joe Zydlewski.
13. Plans for the Future.
    The collection and study of cod sera and ovary data will be continued for the coming year until the accumulated data and analysis can be shaped into a published report on the insights that this approach may offer. It is possible that another live cod will be maintained at the NMF Aquarium Facility at Woods Hole in order to test for the rate of turnover in Vg after induction by estradiol. This would shed light on the meaning of Vg titers during the spent and resting phases of the maturation cycle. There was also discussion with the NMF staff at Woods Hole about possible future extensions of this approach including the possibility that another antiserum to the chorion proteins might shed further light on the late Developing stage when the liver secretes chorion protein precursors into the blood stream.  This later project would require additional funding.

Morrison, C. M. 1987. Histology of the Atlantic Cod, Gadus morhua: An Atlas. Part three, reproductive organs. Dept. Of Fisheries and Oceans, Ottawa, Ontario, Canada.

Respectfully submitted,

Joseph G. Kunkel

jgk/hs