OUP user menu

Prokaryotes and the input of polyunsaturated fatty acids to the marine food web

David S. Nichols
DOI: http://dx.doi.org/10.1016/S0378-1097(02)01200-4 1-7 First published online: 1 February 2003


The investigation of prokaryotes in aquatic ecology is often limited to their role in nutrient cycling and the degradation of organic matter. While this aspect of the microbial loop is undoubtedly important, further aspects of bacterial roles in marine food webs exist which have not been fully considered in light of recent research in related fields. The concept of bacteria providing essential nutrients may derive importance from two aspects of their role in the marine environment; firstly as a primary food source for omnivorous, sestonivorous and filtering benthic animals and secondly as components of the commensal microbial communities of marine animals. Many marine organisms lack the de novo ability to produce n-3 polyunsaturated fatty acids (PUFA) and hence rely on a dietary supply of PUFA. The issue of PUFA origin in the marine food web is particularly salient in light of recent research demonstrating the influence of PUFA levels on the efficiency of energy transfer between trophic levels. The assumption that microalgae provide the bulk of de novo PUFA production for all marine food webs must be actively reviewed with respect to particular microbial niches such as sea ice, marine animals and abyssal communities.

  • Polyunsaturated fatty acid
  • Shewanella
  • Colwellia
  • Deep-sea
  • Piezophile

1 Introduction

Questions regarding the transfer of energy between trophic levels and the limitation of essential nutrients in aquatic food webs have continued to occupy researchers worldwide. Within the wider scope of aquatic ecology, the investigation of prokaryotes is often limited to a consideration of their role in nutrient cycling and the degradation or remineralisation of organic matter. While this aspect to the microbial loop is undoubtedly important, further aspects of bacterial roles in marine food webs exist which have not been fully considered in light of recent research in related fields. In particular the concept of bacteria as potential providers of essential nutrients, such as B-complex vitamins [1], has been raised in the literature from time to time, more recently in relation to the utilisation of bacteria as a food source in aquaculture-based food chains [2]. The concept of bacteria providing essential nutrients may derive importance from two aspects of their role in the marine environment: firstly as a primary food source for omnivorous, sestonivorous and filtering benthic animals [3] and secondly as components of the commensal microbial communities of marine animals.

The available information concerning the phylogeny and distribution of marine bacteria that produce n-3 polyunsaturated fatty acids (PUFA) has been reviewed recently [4,5]. In summary, the ability to produce PUFA exhibits a phylogenetic linkage centred on two distinct lineages, the marine genera of the γ-Proteobacteria (Shewanella, Colwellia, Moritella, Psychromonas, Photobacterium) and more limited species within two genera of the Cytophaga-Flavobacterium-Bacteroides (CFB) grouping (Flexibacter, Psychroserpens). The majority of the γ-Proteobacteria PUFA producers are characterised as being psychrophilic, halophilic and predominantly piezophilic or piezotolerant [5,6] while those of the CFB grouping are similarly psychrophilic and halophilic but lack the ability to grow at high pressure [7]. These physiological traits have influenced the ecological distribution of PUFA-producing bacteria in the marine environment.

Having stated that there appears a strong correlation between the ability to produce PUFA and the phenotypes of psychrophilic and/or piezophilic growth (Table 1), it has been tempting to link their physiological functionality to thermal and/or pressure adaptation of the bacterial cell membrane [5]. However Allen et al. [8] successfully demonstrated the growth of a PUFA-minus mutant of Photobacterium profundum under high-pressure and low-temperature conditions. Hence while PUFA production appears as a phylogenetically linked genotypic strategy for such selective pressures, their presence may not be essential for the growth of bacteria in such environments.

View this table:
Table 1

Major bacterial genera responsible for the production of PUFA in the marine environment

SpeciesPsychrophilicHalophilicPiezophilicPUFAEnvironmental source
S. algaeRed algae, Japan
S. amazonensis+Water, Amazon River
S. balticaOil brine, Japan
S. benthica++++Holourithan intestine
S. colwelliana+Aquaculture, USA
S. frigidimarina−/++Sea ice, Antarctica
S. gelidimarina++−/++Sea ice, Antarctica
S. hanedai++−/++Sediment, Arctic
S. japonica−/++Sediment, mussels
S. livingstonensis−/+ndSea water, Antarctica
S. oneidensisLake Sediment, USA
S. pealeana−/+++Squid gland
S. putrefaciens OGIButter, UK
S. putrefaciens OG3Butter, UK
S. woodyi−/++Sea water, Hawaii
S. violacea++++Deep-sea
C. demingiae++nd+Sea ice, Antarctica
C. hadaliensis+++ndDeep-sea
C. hornerae++nd+Sea ice, Antarctica
C. maris++nd+Sea water, Japan
C. psychroerythraea++nd+Flounder eggs
C. psychrotropica++nd+Burton Lake, Antarctica
C. rossensis++nd+Sea ice, Antarctica
M. japonica++−/++Deep-sea
M. marina++−/++Sea water
M. viscosus++−/+ndFish
M. vavanosii++++Deep-sea
P. antarcticus+++ndSea ice, Antarctica
P. kaikoae++++Deep-sea
P. marina++++Seawater
Ps. gondwanensendBurton Lake, Antarctica
Ps. toquis++nd+Sea ice, Antarctica
Ph. profundum++++Deep-sea
  • nd = not determined.

  • −/+= psychrotolerant or piezotolerant, respectively.

  • This genus contains five other marine-related species. However non produce PUFA.

  • For references see Nichols and McMeekin [4].

  • Bozal et al. [49].

  • Yumoto et al. [51].

  • Nogi and Kato [53].

  • Ivanova et al. [48].

  • Nogi et al. [50].

  • Deming et al. [52].

  • Mountfort et al. [54].

  • Kawasaki et al. [55].

  • Nogi et al. [56].

  • Nogi et al. [57].

2 PUFA: an essential marine food web nutrient

Many marine organisms lack the de novo ability to produce PUFA and hence rely on a dietary supply of n-3 PUFA such as eicosapentaenoic acid [EPA; 20:5(n-)3], docosahexaenoic acid [DHA; 22:6(n-)3] or closely related C18 precursors [2,9]. The effect of this dependence has been demonstrated in a recent study where the level of food web EPA was of central importance in controlling the efficiency of energy and biomass transfer at the pelagic producer–consumer interface [10]. Microalgae have long been considered the major de novo producer of PUFA in marine food webs [11] although further sources such as the lipid-rich thraustochytrids have been suggested to make an appreciable contribution [12]. The ability of marine bacteria to also produce PUFA is now well established [5], although the reiteration of superseded information still occurs in some studies [13]. Given the importance of PUFA in the marine food chain, a possible involvement of bacteria in PUFA provision was postulated [14]. However very little concerted information has appeared in the intervening 16 years to establish the distribution and abundance of PUFA-producing bacteria in the marine environment. It is therefore impossible to assess the overall contribution that bacteria may make to pelagic PUFA fluxes at the present time. Hence the question of prokaryotic PUFA input into marine food webs requires significant clarification in marine ecology and biogeochemistry.

2.1 Deep sea

The abyssal environment is an interesting case in terms of the de novo production of PUFA and transfer to higher trophic levels. While benthic fauna are characterised by containing high levels of PUFA [15], the dietary provision of PUFA from primary production is an area of conjecture. Phytodetritus and faecal pellets represent the most common carbon inputs into abyssal environments and while originally rich in PUFA there is good evidence that in most cases this is removed by microbial activity during the time required for particles to fall through the water column [16]. Likewise particle-associated populations of flagellates, which are dominant in surface waters and may provide a PUFA input to the abyssal region, appear to decline and are replaced with bacterial communities during the deposition process [17]. Available evidence suggests that these bacterial communities are dominated by populations of the CFB group [18] and therefore more unlikely to be a source of PUFA input. However there are suggestions that phytoplankton derived fatty acids can reach abyssal areas under circumstances where the phytodetrital particle size is sufficient to allow a more rapid rate of sinking [19]. Hence it would appear that the biosynthetic origin of PUFA in the deep-sea may be the ability of abyssal animals to biosynthesise their own PUFA or a surface independent microbial community acting as ‘primary’ providers.

DeLong and Yayanos [14] identified piezophilic bacteria with the ability to produce either EPA or DHA from deep-sea sediments and first suggested that these organisms may play an important role in providing PUFA into the marine food web. Subsequently, the linkage between piezophilic ability and the marine related genera of the γ-Proteobacteria, which also produce PUFA, was highlighted [20,21]. However relatively few studies addressed the ecological distribution of psychrophilic and/or piezophilic bacterial populations in the deep-sea. A recent study showed the metabolic activity of natural bacterial communities from abyssal sediment remained higher when samples were incubated at in situ versus atmospheric pressure. This was interpreted to indicate the dominance of a piezophilic community [22]. Further, piezophilic isolates (Shewanella and Moritella spp.) dominated mixed cultures obtained from incubations of abyssal sediment under in situ pressure. In contrast control incubations at atmospheric pressure were dominated by cosmopolitan marine Pseudomonas spp. [23]. Interestingly Kaneko et al. [24] isolated a piezotolerant Pseudomonas spp. from the Japan Trench (4418 m depth). However the isolate possessed a maximum growth pressure (MGP) of between 40 and 50 MPa at 10°C and a decreasing MGP with temperature between 30°C and 10°C. It therefore seems unlikely this bacterium would be able to grow under the in situ conditions from which it was isolated (2–4°C and 44 MPa). Such studies provide arguments for the existence of a functionally dominant piezophilic community in abyssal sediment which could provide a de novo source of PUFA for filtering benthic animals.

However doubts as to the in situ metabolic activity of abyssal bacterial sediment communities challenge this hypothesis. Piezophilic (and PUFA-producing) bacteria may be classed as copiotrophic, with the possibility that their metabolic activity may vary with nutrient concentration [25]. Many studies of the metabolic activity of abyssal sediment communities have employed the addition of unnaturally high levels of nutrient and/or labelled substrate, raising the concern that an artificially high metabolic activity may have been induced. This concern was highlighted by Wirsen and Molyneaux [26] who found that growth rates of natural deep-sea populations were very low or below detection when incubated in a high pressure chemostat with naturally occurring concentrations of seawater carbon. However the addition of supplemental carbon at low levels did result in a positive growth response. Similarly, piezophilic growth has been demonstrated in studies using labelled carbon substrate at concentrations less than 10% above natural levels [27]. The implication is, therefore, that while microbial populations introduced into abyssal regions through phytodetritus fall are metabolically inhibited by high pressure (i.e. CFB and Pseudomonas spp.), the metabolic activity of the piezophilic populations present in abyssal sediment is limited by the arrival of phytodetrital nutrients.

The limited metabolic activity of abyssal sediment bacteria is also supported by microscopic and biomarker studies. While bacterial cell numbers in surface sediment appear to be around 2–3×108 cm−3 from the Porcupine Abyssal Plain region of the North Atlantic [22], a lipid biomarker study from the same region did not detect any PUFA in surface sediment samples. The detection of PUFA would be expected if an active, dominant community of piezophilic bacteria were present. Branched-C15 fatty acids were detected in appreciable amounts from one of five samples, which may have been indicative of piezophilic Shewanella spp. However again no PUFA was reported from this sample, which would have been expected based on the fatty acid composition of cultured piezophilic Shewanella spp. which produce almost a 1:1 ratio of branched-C15 fatty acids:PUFA [6]. The non-detection of PUFA from abyssal sediments is a common result from diverse areas [16] and appears the best argument to support the limited growth and activity of cultivated piezophilic bacteria in abyssal sediment.

Hence what is the optimal ecological niche of piezophilic PUFA-producing bacteria and where are they likely to provide a significant contribution to food web PUFA? Abyssal sediment communities appear unlikely to be a major direct dietary source. However they may act as an inoculum for the intestinal communities of abyssal filter feeders where they may grow with the benefit of higher nutrient concentrations [28]. There is good evidence for the existence of stable intestinal communities in abyssal holothurians and the frequency of PUFA-producing piezophiles isolated from the intestinal contents and lining of such animals suggests a resident community [28,29,30]. In the context of the deep-sea it may be that this environment represents the favoured ecological niche for piezophilic PUFA-producing populations. The assimilation of lipids within the intestines of abyssal holothurians is supported by the finding of ester hydrolase and lipase activity within tissues of the anterior and posterior intestine where microbial communities are suggested to be of similar structure to that of the surface sediment [30]. Hence the assimilation of PUFA from intestinal bacterial communities is a distinct possibility. The demonstration of one ecosystem that may rely on PUFA-producing bacteria raises the question of whether bacterial PUFA input in other ecosystems is also more significant than previously considered.

2.2 Sea ice

One habitat where PUFA producers represent common culturable isolates is the annual sea ice surrounding Antarctica. Sea ice provides one of the major niches for microorganisms in the Antarctic region and critically influences the productivity of the Southern Ocean [31]. During the austral spring and summer sea ice supports the growth of a wide array of microalgae, mostly diatoms, found primarily near the bottom of the hard congelation ice [32,33]. Microalgae may contribute up to 50% of primary production in certain regions [34] and produce arachidonic acid [20:4(n-)6] in addition to EPA and DHA. Ice-associated bacteria are responsible for most of the secondary productivity in sea ice [35]. The prokaryotic community appears to be dominated by psychrophilic populations of both free-living (mostly γ-Proteobacteria) and epiphytic (mostly the CFB group) types [36,37] although communities can be heterogeneous in nature, differing markedly at the genus/species level between ice samples [38]. A high proportion of the psychrophilic taxa isolated from sea ice possess the ability to produce PUFA and have included many new species closely related to those isolated from deep-sea sediments (Table 1). However no information is yet available on the abundance of PUFA-producing bacteria within the populations of psychrophilic bacteria that dominate the prokaryotic sea ice community from Antarctica. From the Arctic the phylogenetic diversity of numerically important sea ice bacteria has established the presence of PUFA-producing bacterial populations, including one isolate with a 100% 16S rDNA sequence homology to the Antarctic PUFA producer Shewanella frigidimarina [39]. This suggests that PUFA-producing bacteria may play an important role in the food web of global polar marine ecosystems.

2.3 Marine animals

PUFA-producing bacteria have also been commonly isolated from the intestinal contents of marine fish and invertebrates. A comparison of 7000 bacterial isolates from the intestinal contents of temperate fish, zooplankton, shellfish and surrounding seawater/sediment yielded 112 PUFA-producing strains. The percentage of positive isolates from different host species varied, with a maximal value of 2.9%. Interestingly no positive strains were isolated from the associated seawater or sediment samples [40] and led the authors to consider a specific linkage between PUFA-producing bacteria and intestinal communities. From polar regions a similar implication is supported. The investigation of culturable isolates from 10 Arctic and sub-Arctic invertebrates revealed PUFA-producing bacteria from all organisms. In total 98 from 258 isolates produced EPA or DHA (38% average) rising to 50% of isolates for two bivalve and one amphipod species. The authors concluded that PUFA-producing bacteria were a dominating element of the culturable bacteria in some invertebrates [41]. Further evidence suggests that PUFA-producing populations are also distributed through the intestinal communities of higher marine organisms. It was estimated that DHA containing bacteria accounted for 14% and EPA-producing bacteria 30% of total cell counts in the intestinal contents of seven deep-sea fish. In contrast no DHA-producing isolates were found from 112 cultures of intestinal bacteria from 10 shallow-sea animals. However EPA production was found for 40 isolates [42]. It was concluded that bacteria containing PUFA were indigenous to the intestines of deep-sea fish and are generally distributed in deep-sea environments. Further it was considered that PUFA-producing bacteria actively grow in the intestines of deep-sea fish and account for a large proportion of the bacterial community. These conclusions are supported by similar studies [43,44].

The importance of PUFA-producing bacteria to marine animals is circumstantially supported by the discovery of symbiotic relationships. Shewanella isolates phylogenetically grouped with the PUFA-producing species S. gelidimarina, S. benthica, S. hanedai and S. pealeana were associated with the accessory nidamental glands and egg capsules of squid specimens (Loligo pealei) where they formed symbiotic populations within a larger bacterial community. Similar isolates were identified from squid species in both the Atlantic and Pacific oceans [45]. It was considered that the bacterial communities are integrated into the reproductive life cycle of the squid although their role is unclear. The chemical defense of egg predation was suggested through the production of toxins (e.g. tetrodotoxin). PUFA production, an essential nutrient for marine larval development, may be a further role worthy of consideration.

While the production of PUFA as cell membrane components by bacteria in the intestinal environment of certain marine animals may be documented, the potential benefit of this association must be considered in order to establish a hypothetical rationale for their ecological role in the marine food web. Again, environmental studies in this area are lacking. However evidence for the nutritive value of PUFA-producing bacteria does exist from aquaculture studies. Phillips [1] summarised a number of studies detailing the enhanced survival and growth of detritovors when fed on microbially enriched detritus; the implication being that bacterially derived vitamins or other micronutrients may have been responsible. However at that time Phillips [1] considered that bacteria would be a poor food source for marine organisms due to their perceived lack of PUFA. The transfer of bacterially derived fatty acids, and specifically PUFA, between marine bacteria and higher trophic levels has been demonstrated [46,47]. The question of ecological significance, still unanswered, is whether such transfer occurs in the intestinal environment of marine organisms harbouring PUFA-producing bacterial populations. If so, it could be envisaged that certain fish possess a symbiotic relationship with intestinal PUFA-producing bacteria which would provide them with a baseline supply of PUFA nutrients which are essential and growth limiting.

3 Conclusions

The issue of prokaryotic PUFA production is particularly salient in the light of recent research which has demonstrated that EPA production may significantly influence the efficiency of energy transfer between primary and consumer trophic levels in aquatic ecosystems [10]. The assumption that microalgae provide the bulk of de novo PUFA production for all marine food webs must now be actively reviewed to determine the role and potential importance of PUFA-producing bacteria within certain marine food chains.

The barrier to addressing the quantification of prokaryotic PUFA production in the marine environment has been the absence of classical methodologies for the identification and enumeration of PUFA-producing bacterial populations. Recent developments in molecular microbial ecology and the genetics of PUFA biosynthesis now allow molecular methods to be used in this area. The application of techniques such as fluorescent in situ hybridisation-microautoradiography and stable isotope probing combined with biomarker approaches offer significant advances to understand the role of PUFA-producing prokaryotes in marine microbial niches such as sea ice, marine animals and abyssal communities.


Drs. Kevin Sanderson and John Bowman are thanked for comments on the manuscript and Prof. Jody Deming for helpful conversations.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
  52. [52].
  53. [53].
  54. [54].
  55. [55].
  56. [56].
  57. [57].
View Abstract