LIFE IN THE OCEANS: TEMPORAL AND SPATIAL DISTRIBUTIONS
Marine organisms, in general, have not been studied as extensively as their terrestrial counterparts, although studies of marine organisms have made major contributions to our fundamental understanding of ecological factors, such as keystone species, predation, and competition. Several fundamental questions underlie much of the research in biological oceanography today: What are the spatial distributions of marine organisms? How do these distributions vary with time? What are the relationships between distribution and function (i.e., production, metabolism, material cycling, and flux)? What are the mechanisms that drive these processes, including biological interactions, physical and chemical influences, and physiological responses of organisms?
Marine biologists have led the study of ecological interactions for several decades, but molecular techniques have not been widely used in this field. Molecular methods can allow us to proceed beyond descriptions of population structures to discern the mechanisms that govern population regulation, migration, and distributions. Understanding the control of spatial and temporal distributions of marine organisms will become increasingly important as we attempt to determine how natural and human-caused environmental changes affect population sizes, extinction, and invasion of new species. Although the need for such information is more obvious for species on which we depend directly for food or other uses, other species perform vital, if sometimes obscure, roles in ecosystems.
For the majority of marine species the abundance, genetic diversity, and spatial limits of populations have not been determined or are poorly understood. Knowledge of species distributions and genetic structure are important for (1) management of commercially important fisheries, (2) conservation and protection of endangered and threatened species, (3) assessment of the impact of global change and pollution on natural populations and the ocean's impact on global processes, (4) creation of models that describe and predict global energy flow and biogeochemical cycles, and (5) determination of the mechanisms of extinction and speciation in ancient and modern oceans and the resulting marine biodiversity.
Remarkable advances in our understanding of the problems outlined above have been achieved in the past decade by exploiting molecular biological techniques and approaches. The genetic variability of some populations of fish, cetaceans, sea turtles, phytoplankton, zooplankton, and other marine organisms has been revealed (reviewed by Falkowski and LaRoche, 1991; Powers, 1993). These studies demonstrate the promise of molecular techniques for answering questions that were previously intractable. Studies examining the genetic variability of the stocks of several species of commercially harvested fish e.g., (salmon, tuna, pollack, blue marlin) have demonstrated that new mitochondrial and nuclear DNA technologies will permit a definition of population structure, identification of “stocks,” and quantification of gene flow. Such information is essential for effective management of oceanic fish populations.
Some fisheries, especially in North America, are declining, while others cannot be accurately accessed because of a lack of scientific data (U.S. Department of Commerce, 1992). Strategies such as the recent move toward moratoria on salmon fishing in the Pacific and swordfish fishing in the Atlantic can be avoided only if multinational management efforts are successful. Such management depends on knowledge, for each fishery, of the source (i.e., spawning grounds) of each fish, population sizes and genetic diversity that could produce a sustainable harvest, and the nature and extent of gene flow among separate populations, which may be critical to their long-term stability.
The ability to characterize genetic features of key marine species will illuminate the biology of these organisms by providing insight about their life history patterns, breeding structure, and the transfer of genetic information among populations. This information can also be compared with information about the variability of species abundance, ecosystem structures, and environmental conditions, allowing scientists to determine correlations and causal relationships among variables and to improve the models and techniques used by fisheries managers. In recent years molecular genetic characterization of endangered and threatened species, including cetaceans and sea turtles, has allowed better definition of population structures, genetic diversity of populations, and variability
in migration routes (e.g., Baker et al., 1990; Bowen and Avise, 1990; Meylan et al., 1990; Amos et al., 1991, 1992).
Molecular techniques also offer great promise for the study of microbial populations. Thousands of different species of phytoplankton, bacteria, fungi, and viruses existing in marine ecosystems comprise complex microbial food webs. These organisms are highly diverse in terms of the ecological and biogeochemical roles they play in elemental cycling and exchange within and between the ocean and atmosphere. Most of these species have not yet been named because they are not distinguishable morphologically from one another and cannot be cultured for taxonomic, ecological, and physiological studies.
Using RNA and DNA sequences to identify microbial groups, a data base describing the microbial diversity at the genetic level has been developed (Fox et al., 1977, 1980; Woese, 1987). Figure 2 shows one of the first phylogenetic trees determined from rRNA sequence comparisons. The genetic data base is being expanded rapidly as new sequences are added. In combination with phylogenetic probing and DNA sequencing, detection of functional genes that are similar to genes of well-studied species can help characterize these largely undescribed microbial species with respect to their ecological and biogeochemical roles in the marine environment. Because some species are difficult to culture, it is often easier to describe the genetic structure of organisms than to describe the organisms themselves. For example, DNA sequences that are uniquely found in archaebacteria, a primitive group of bacteria previously thought not to occur in the surface of the oceans in appreciable numbers, have been obtained from seawater samples (e.g., Giovannoni et al., 1990). Furthermore, this approach has revealed the diversity and uniqueness of nonculturable symbiotic bacteria (e.g., Haygood and Distel, 1993). Use of this approach to study the diversity of important marine taxa other than microbes is becoming common (e.g., Palumbi and Benzie, 1991; Bucklin et al., 1992; Finnerty and Block, 1992; Silberman and Walsh, 1992).
Understanding the population dynamics of marine invertebrates has been hindered by our inability to distinguish among larvae of different species. Larval recruitment is a key event in the life cycle of most marine invertebrates and in the population dynamics of organisms that feed on them. Molecular taxonomic techniques have the power to advance our understanding of the ecology of these species greatly. In addition to facilitating studies of population structure and dynamics in marine organisms, molecular techniques have great potential for assessing the physiological status of individual marine organisms. In bacteria, for example, the amount of ribosomal RNA in the cell is related to growth rate and can be easily assayed by using fluorescent molecular probes. Finally, for photosynthetic organisms, the amount of the enzyme RuBP carboxylase in a particular cell can be an indicator of photosynthetic potential.
Recent research has demonstrated that viruses are abundant in the ocean, being present at levels of between 10,000 and 10 million per milliliter of seawater (e.g., Proctor and Fuhrman, 1990, 1991). Little is known about their relationships with their host organisms, and thus the role they play in regulating the population dynamics of microorganisms in the sea is obscure. Because of the abundance of viruses, it has been suggested that viral infections may play a central role in marine bacteria and phytoplankton mortalities and in maintaining the diversity of microbial populations (Suttle et al., 1990). Our understanding of the biology of the ocean may be altered dramatically when the role of viruses is understood, in much the same way that our understanding of the human “ecosystem” was greatly enhanced after the relationship between viral infection and disease was established. Marine viruses offer tremendous potential for biotechnology as shuttle vectors, providing a vehicle to move selected pieces of DNA into a target organism.
Molecular tools could be used to study a variety of ecological processes, such as competition, predation, parasitism, and coevolution of host and parasites and predators and prey. The mechanisms that marine organisms use to select habitats are also an important topic that is just beginning to be addressed.
Techniques to Address the Scientific Questions
Several molecular methods have been developed for the analysis of proteins, RNA, DNA, and metabolites that can be applied to study how individuals, populations, and communities are distributed and to begin to assess the physiological status of individual organisms. The methods and techniques include (1) isozyme analysis; (2) immunochemical methods; (3) nucleic acid hybridization techniques; (4) restriction fragment length polymorphism (RFLP) analysis of mitochondrial, chloroplast, and/or genomic DNA; (5) DNA amplification by PCR; (6) DNA fingerprinting; and (7) DNA sequence analysis.
Isozyme Analysis: For almost three decades, electrophoresis of cell extracts followed by histochemical staining to identify specific protein electromorphs, has uncovered a wealth of genetic variations at the molecular level, both within and among species (Figure 3). These time-proven methods have continued to be useful to distinguish among species, analyze transport processes in the ocean, determine the genetic architecture of natural populations, study hybridization among species, and study the adaptive significance of isozyme variation in natural populations (e.g., Bucklin et al., 1989; Buth, 1990).
Immunochemical Methods: Immunochemical methods can detect species-or population-specific molecules (antigens) on the surfaces of cells or in cell extracts and have been used to identify bacteria, phytoplankton, zooplankton, eggs, larvae, and adult marine organisms. One application of this technique has been for identification of aquatic nitrifying bacteria (Ward and Perry, 1980) and nitrogen-fixing microorganisms (Currin et al., 1990) and mapping of their distribution over large areas. In addition to the identification of species and life history stage, immunochemical methods have been used to describe community structure and predator-prey interactions, to identify the partially digested remains of prey organisms from within the digestive systems of their predators (Theilacker et al., 1986; Grisley and Boyle, 1988; Feller et al., 1990), and to detect parasites (Ohman et al., 1991).
Nucleic Acid Hybridization: Nucleic acid hybridization, like isozyme analysis, can be used to discriminate among organisms that lack distinguishing morphological features, such as microorganisms and larval forms of complex multicellular organisms. DNA hybridization can be designed to discriminate among major groups of organisms or to detect individual species within a group of closely related organisms (Woese, 1987). Probes that target unique DNA sequences are prepared and reacted with DNA from the organisms of interest. This technique has already been used to identify species in marine microbial populations (Delong et al., 1989; Sanghoon and Fuhrman, 1990).
RFLP Analysis: RFLP analysis is used to characterize the genetic composition of organisms. Digestion of genomic DNA, mitochondrial DNA, or chloroplast DNA by enzymes that target specific DNA sequences (restriction enzymes) generates DNA fragments that can be separated by electrophoresis to form distinctive patterns that can then be analyzed by standard statistical techniques. Comparison of the similarity among patterns has been used to identify the geographic origin of individuals and to study the taxonomic relationships among species. Information gained by RFLP analysis is especially useful in fisheries management, conservation of endangered species, transport studies of plankton, and studies of evolutionary relationships. RFLP analyses have also been used to study the genetic diversity of a variety of marine and freshwater organisms, including bacteria, marine macrophytes, phytoplankton, zooplankton, eels, oysters, sea turtles, and fish.
Polymerase Chain Reaction (PCR): This technique has provided a revolutionary approach for the synthesis, detection, and characterization of specific sequences of DNA. The PCR procedure employs a heat-stable
bacterial DNA polymerase and specially constructed oligonucleotide primers to replicate specific DNA sequences in vitro. From as little as a single molecule of a target DNA sequence, enough material for standard analytical procedures such as RFLP analysis, gene mapping, DNA hybridization, DNA fingerprinting, or even DNA sequencing can be produced in about three hours. Advantages that can be gained by the use of PCR include (1) rapid detection and identification of microorganisms that occur at very low frequency (e.g., one or a few cells in a liter of water), symbiotic microorganisms, and minute individual fish and invertebrate larvae; (2) rapid analysis of individual genomes for population studies; (3) detection and analysis of “rare events” (e.g., gene rearrangements) that occur in a small fraction of cells in a tissue sample or field collection; and (4) estimation of water quality by detection of pathogenic viruses, bacteria, and/or parasites. Another advantage of PCR is that it eliminates the need to culture what are often found to be nonculturable microbial species and strains of ecological or biogeochemical importance. PCR currently is being used for biosystematics, population biology, conservation biology, ecology, developmental biology, and genetics (see reviews by Arnheim et al., 1990; Powers et al., 1990; Powers, 1993).
DNA Fingerprinting: Repetitive regions of DNA, called minisatellites, are dispersed throughout the genomes of a number of organisms. Jeffreys et al. (1985) showed that a subset of human minisatellites shared a common 10 to 15 base pair core that had hypervariable regions. Later they demonstrated that a nucleic acid hybridization probe could detect highly polymorphic minisatellites that could be used as DNA fingerprints specific to an individual. DNA fingerprinting is now commonplace in biomedical research and is routinely employed in a variety of legal situations. Moreover, the method is being used to address many scientific questions about terrestrial and aquatic organisms (reviewed by Ryland and Tyler, 1989; Burke et al., 1991; Powers, 1993).
Whales and other cetaceans have been studied more frequently than other marine species (e.g., Hoelzel et al., 1991; Schloetterer et al., 1991). However, DNA fingerprinting has also been applied to freshwater and marine fishes, invertebrates, and aquatic plants (reviewed by Ryland and Tyler, 1989; Powers, 1993). For example, Whitmore et al. (1990) used DNA fingerprinting to study sibling largemouth bass (M. salmoides.) They showed that DNA hydridization patterns were different for each individual but that siblings were more similar to each other than to fish from wild stocks. Wirgin et al. (1991) developed striped bass-specific DNA probes, 10 to 20 base pairs in length, which they used to study bass population structure. One of the probes allowed them to distinguish between Gulf of
Mexico and Atlantic coast populations as well as discriminate among several of the Atlantic stocks. Turner and his colleagues (1991) have used repetitive DNA sequences to study the population structure of several fish species.
Several laboratories are using a PCR-based approach to amplify DNA sequence length polymorphisms that have DNA repeats of specific nucleotide combinations. Such approaches can be used to examine hundreds of individuals rapidly, at a modest cost. Moreover, the approach can be used for fresh, frozen, or alcohol-fixed samples from a broad array of geographic areas and from any particular life history stage (eggs, larvae, fry, or adult). Data generated by DNA fingerprinting techniques can be used for assessment of genetic stocks, for restoration and mitigation activities, and for the management of critical estuarine and marine resources.
DNA Sequence Analysis: DNA sequence analysis, in which the actual sequence of the DNA subunits (nucleotides) is determined, is used to study genetic relationships within and among marine communities, species, and populations. When PCR amplification is used to provide the DNA for analysis, the sequence information can be obtained without sacrificing the organism or, in the case of microorganisms, without culturing them in the laboratory. Nucleotide sequences of specific genes are also used to establish taxonomic relationships, to study the genetic architecture of natural populations, to manage endangered species, and to study the evolution of marine organisms. During the past five years, there has been an increase in these types of studies in the marine science community (reviewed by Avise, 1993; Powers, 1993).
Automation of Molecular Techniques
The efficiency and usefulness of molecular techniques are increased in those instances in which automation expedites the processing of large numbers of biological samples. Automation is important for processing samples for molecular analysis because it is often necessary to collect a large number of samples from a given study area. Large sample numbers when processes that vary greatly over time or space are being studied and/or when repeated sampling at the same site is required. A number of methods for protein and nucleic acid analysis have already been automated, and others are in the developmental stages as a result of the Human Genome Project and their commercial applications for diagnostic purposes in the biomedical field.
For many types of biological analyses, samples must be collected and preserved in the field and analyzed later in the laboratory. Automated techniques to replace human identification and enumeration of microbes, invertebrate larvae, and other organisms not readily identified on the basis of gross morphology could make near-real-time analysis of biological samples at sea or in the field a reality. A few of these automated methods are highlighted below.
Protein Isolation: Automation of methods for protein isolation is the paramount problem in characterizing proteins from complex living organisms, organelles, or even cells, requiring the separation of thousands of different proteins. The technique best suited for this type of analysis is two-dimensional (2D) gel electrophoresis, whereby tens of thousands of proteins may be separated in one dimension by size and in the second dimension by electrical charge. Computerized image analysis is presently used to digitize information from 2D gels so that it may be stored in the computer. This process allows comparison of different 2D gels to determine the qualitative or quantitative differences in the locations of protein spots. With this technique it will be possible to carry out, on a routine basis, 2D gel electrophoresis and subtractive analyses of marine organisms or cells. This type of analysis will be even more useful when fully automated 2D gel techniques are developed.
Once appropriate proteins have been identified for characterization, a technique termed “electroblotting” can be used to transfer proteins from 2D gels directly onto appropriate materials from which their amino acid sequences may be determined.
Protein Sequencing: Automated methods have been developed for sequencing proteins, allowing the determination of protein sequences at the subpicomole level. The current state-of-the-art sequencing technique, using a conventional gas-phase protein microsequencing instrument, permits the analysis of approximately 10 picomoles of material in a typical sample. New solid-phase sequencing instruments and techniques are currently being developed to permit the use of fluorescent phenylisothiocyanate (PTH) derivatives, enhancing the detection of PTH amino acid derivatives 10,000-fold and speeding the throughput of protein sequencing fivefold. Within the next few years, fluorescent sequencing and solid-phase techniques will permit analysis of proteins at the level of 100 femtomoles.
A second approach to protein microsequencing uses the mass spectrometer as a tool to isolate peptide fragments for subsequent sequencing. Techniques such as electrospray and plasmid desorption will permit, in the not-too-distant future, analysis of intermediate-size peptides (up to 40 residues) at levels that will probably fall well below the picomole level. The power of the mass spectrometer as a tool for analyzing peptides is that it can characterize peptides rapidly (in
minutes) and identify, in a general way, any secondary modifications that have occurred on proteins. After proteins are sequenced, standard techniques can be used to identify the genes corresponding to these protein products.
Both solid-phase and mass spectrometer approaches allow rapid processing, with cycle times that are as little as 15 to 25 percent of the present sequencing times. Thus, these combined techniques will allow many proteins from marine organisms to be characterized and their corresponding genes to be isolated.
DNA Mapping and Sequencing: Automated DNA mapping and sequencing methods have been developed that make DNA sequencing rapid, reliable, and accessible to a broader range of scientists. Fluorescent DNA sequencing technology allows analysis of between 40,000 and 60,000 bases per week by a single technician operating one machine. This output represents an enormous increase in the capacity for processing DNA sequences, so that genes from a large number of samples can be characterized on a greatly increased scale suitable for population studies.
New techniques for both genetic mapping and physical mapping are being developed to increase throughput and increase their potential for automation. For example, techniques are being developed whereby four different fluorescent dyes can be applied to DNA physical mapping. Because four different emission wavelengths can be used, three unknowns and one standard can be run in each lane of a pulsed field electrophoresis gel. Another technique, the oligonucleotide ligase assay, permits individual base polymorphisms to be analyzed. This technique has been automated for DNA mapping work with a robotic workstation.
Computational Tools: The advancement of techniques for analyzing protein and DNA sequences will require appropriate computational tools for the acquisition, storage, retrieval, and analysis of the resulting information. For example, specialized computer chips have been developed recently that have the capacity to detect patterns in DNA or protein sequence data at speeds that are orders of magnitude faster than possible with conventional computers. In addition, these chips can be attached to personal computers or to relatively simple workstations, as well as to larger computers. Other computational techniques, such as neural nets, can be developed to analyze informational patterns. Object-oriented data bases will permit the storage of many different types of data; differences in data type will be transparent to the user. One of the new computational problems in biology is the existence of many different data bases that must be transparent to provide easy access and movement through a variety of data bases to gather the desired information.
Molecular and/or pattern recognition systems (e.g., receptors, functional groups, specific antibodies, enzymes, and metal-binding proteins) can be associated with a monitor in such a manner that they generate an extensive array of molecular recognition elements, coupled via a series of signal transducers. Binding of specific molecules to these recognition elements causes a conformational change in the receptor molecule that is sensed and amplified by the transducer. Thus, the binding of specific molecules found in organism extracts to a specific recognition element can be detected. Such systems have the potential for accommodating a vast number of detector molecules or receptors. The potential of the system is presently limited by the dearth of fundamental biological information on specific molecules and their receptors in marine environments. As this information becomes available, coupling the recognition elements to an automated system could be easily accomplished and would permit rapid assessment of their presence in biological extracts or water samples.
Flow Cytometry: Flow cytometry, which has been used for cell analyses in biomedical research for the past 20 years (Shapiro, 1988), has been adapted for the study of marine organisms (Shumway et al., 1985; Chisholm et al., 1986, 1988; Olson et al., 1990). A new generation of multichannel flow cytometers is being developed with the capacity to identify and count an array of planktonic species automatically, as part of a shipboard multiinstrument water sampler. Submersible versions are in the design stages. Flow cytometers are being used at sea to separate and count selected species of plankton by either their natural fluorescence emission or their size or by detection of fixed cells with molecular probes possessing fluorescent reporters. This approach has expanded the ability to determine the spatial distributions of marine microbes on a scale previously impossible even with conventional microscopy. Applications of this analytical approach include water-quality monitoring, rapid and quantitative detection of indicator organisms, and monitoring of the distribution and abundance of selected microbial species in polluted estuaries and harbors by the enumeration of different metabolic groups of bacteria (e.g., sulfate reducers, denitrifiers, nitrogen fixers, nitrifiers, methanogens, or other microbes), many of which play critical roles in the biogeochemical cycles controlling greenhouse gases.
Microscopic Techniques: New microscopic techniques that afford high resolution can be exploited to determine functional relationships of organisms, to explore development, and to evaluate molecular mechanisms such as ligand-receptor binding. These microscopic techniques include (1) confocal, (2) atomic force, (3) optical force, and (4) scanning Fourier-transform infrared.
The molecular techniques and approaches described above provide a means to address fundamental biological oceanography questions. Although some advances in our understanding of the distribution and function of marine organisms have been made, exploitation of molecular techniques could provide insight into critical issues facing marine science today, including global climate change, management and conservation of living marine resources, seafood quality, and bioremediation.
One of the critical questions facing marine science today is the role of the ocean in global climate change. It is likely that the ocean and atmosphere behave as a coupled system and that changes in one induce changes in the other. Changes in climate could influence oceanic circulation patterns (Broecker, 1991), which could, in turn, change the patterns of productivity and dispersal of organisms in the ocean. Changes in the biological communities of the ocean could significantly alter biogeochemical cycles for elements such as carbon, altering atmospheric concentrations of carbon dioxide and other greenhouse gases. Without a relatively complete understanding of the factors that regulate the population dynamics of existing populations, we will have great difficulty predicting and/or monitoring changes in populations of marine organisms that might be affected by human activities.
Research programs have been initiated to define and describe the ocean's role in global change and the potential effect of global change on marine organisms; molecular biology can play a part in accomplishing the objectives of these programs. For example, a major goal of the Joint Global Ocean Flux Study is to gain an understanding of how carbon dioxide is exchanged between the atmosphere and the surface ocean and how carbon is cycled through physical, chemical, and biological effects and transferred to the deep ocean (Figure 4). Coastal and marginal seas, where greater than 50 percent of primary productivity occurs, are important carbon sinks (Mantoura et al., 1991), and research funded by the U.S. Department of Energy (DOE) and the National Oceanic and Atmospheric Administration could be enhanced by using molecular techniques for studying these processes. DOE has started a new initiative called Molecular Approaches to Ecosystems Research and has used this to augment its Ocean Margins Program, and the Office of Naval Research has a molecular biological component to its new Marine Environmental Quality program. Molecular techniques can be used to provide information about population size and community structure in surface waters and the deep ocean, which is essential to understanding carbon dioxide exchange and carbon cycling. In addition, they will be important in linking key primary producers with mineralizers controlling the net flux of carbon dioxide into coastal waters and its ultimate fate once fixed into complex organic molecules.
The ocean is a source of many resources, including food. As worldwide population grows, the demand for these resources will also increase. Many fish and shellfish populations throughout the world ocean are showing signs of decline due to both harvesting and disease (Figure 5). For example, standing stocks of some commercially important fish (tuna, swordfish, salmon, shark) that live in the ocean have declined sharply in recent years. Fisheries managers usually lack necessary data to demonstrate stock size, the genetic diversity of the population, and the amount of interbreeding among populations in different ocean basins. The degree to which fish populations are controlled by various environmental factors is poorly characterized, although there is some evidence that physical factors produce effects that ripple upward through marine food webs. For example, the Global Ecosystems Dynamics Experiment (GLOBEC) cites evidence that increased winds over the North Atlantic Ocean in the 1950s to 1970s decreased phytoplankton and zooplankton production, reducing the food supply and populations of some fish species (National Research Council, 1991). Information needed to answer questions of stock size and genetic diversity could be gathered by combining data obtained using molecular techniques with a better understanding of the physics and food web structures of ecosystems. Such information could also result in better fisheries management and possibly increased harvests of commercial fish stocks (Figure 6). A strong and productive fishing industry could increase the amount of fish and shellfish exported from the United States. The GLOBEC program will focus on how a changing global environment could alter the stability and productivity of marine ecosystems. GLOBEC planning has placed emphasis on the use of molecular techniques for its studies.