Our previous studies on Hg(II) biotransformation have shown that it can be converted into metacinnabar (HgS) in both eukaryotic algae  and prokaryotic cyanobacteria . However, these studies did not investigate supplementation with sulfur containing compounds, nor did they assess metal sulfide production in response to Cd(II) exposure. Cadmium is also a Group 12 stable metal that is very toxic and widely distributed in the environment.
Tolerance to Cd(II)
The levels of Cd(II) tolerance of the algae and cyanobacteria were similar to previous studies , and the addition of sulfate to all of the investigated species both prior to and during metal exposure resulted in higher tolerance to Cd(II) than any other treatment (Figure 1). These cultures were either the same as (Cyanidioschyzon and Synechococcus) or only slightly lower in biomass (Chlamydomonas) over the 48 h growth period by comparison to the metal-free controls. Although cadmium stress has been shown to induce sulfur limiting conditions [7, 19], this was not entirely alleviated by the simultaneous provision of sulfate in any of the studied species, thus indicating that established metabolic reserves of sulfur other than sulfate itself, may be involved in cellular protection. Furthermore, it has been demonstrated that Cd exposure triggers a decline of photosynthetic apparatus thereby liberating sulfur as well as nitrogen and iron, which can be subsequently used for the synthesis of Cd detoxification enzymes . Assimilated sulfate appears to create an organic sulfur pool that can be readily employed to biotransform Cd(II) as it enters the cell in a similar manner to that proposed for Hg(II) where chemical modification of thiols severely lessened HgS production [14, 15]. Why this cannot be provided by simultaneous sulfate provision is likely to be a product of the high energy demand (732 kJ mol-1) required to reduce sulfate to sulfide for thiol production, energy required for sulfate uptake, and the decline in sulfate uptake induced by cadmium itself . These organisms rely on photosynthesis to generate reducing power that is essential for carbon fixation. If this is shunted towards sulfate assimilation, it would inhibit cellular metabolism and growth. By temporally displacing energy requirements to a pretreatment period, this is overcome and the cells are able to adequately cope with any stress imposed by subsequent exposure to Cd(II). The simultaneous sulfate and metal treated cells grew marginally better than the cells treated with metal alone in Cyanidioschyzon and Synechococcus (Figures 1B & C), but not in Chlamydomonas. Metabolic differences might ac-count for this; i.e. the former species may have relatively more efficient sulfate assimilation. Interestingly, in a separate study it was revealed that Synechococcus is able to utilize elemental sulfur as a sulfur source resulting in enhanced metal tolerance (data not shown). These results point to the importance of sulfur nutrition in cadmium tolerance that has implications for other organisms [20, 21], including humans . Nevertheless, this has not been well documented in the literature.
The other treatment in which Synechococcus grew better than in cadmium alone was that in which cysteine was supplied both prior to and during metal exposure. However, this cannot be accounted for by a relatively high cysteine desulfhydrase activity in Synechococcus (Figure 4). Both eukaryotic species were not as adept at coping with this form of sulfur supplementation.
In general, pretreatments with sulfite, an intermediate of the sulfate assimilation pathway, and with the amino acid, cysteine, did not provide ameliorative effects on growth during Cd(II) exposure to the same degree as sulfate. Sulfite can be toxic to green algae  because of interactions with sulfide bonds of glutathione and glutathione disulfide that severely affect anti-oxidation processes . It can also lead to SO2 toxicity through sulfoxy-free radicals generated by the oxidation of SO3
2- by O2
−. Furthermore, in membrane preparations of cyanobacteria, sulfite stimulates ATP hydrolysis and inhibits ATP synthesis . Exogenous cysteine is believed to have direct effects on transporters and enzymes that are sensitive to thiol/disulfide redox variations . This could account for the deleterious effects on the eukaryotic organisms in this study as unfortunately, these treatments did not improve Cd(II) tolerance. However, cysteine did improve the growth of Synechococcus in the presence of cadmium. It is possible that this organism is not as susceptible to functional interference of its protein thiol groups, or that it has a greater absorption and storage capacity for cysteine, thereby lowering its deleterious effects.
Cellular sulfide production
The measurement of acid labile sulfide is a convenient way to estimate amounts of metal sulfide within samples . Our studies clearly indicated that the addition of Cd(II) caused de novo aerobic synthesis of metal sulfide, assumed to be predominantly CdS because there was no detected increase in metal sulfides when Cd(II) was not supplied to the cells under any conditions (data not shown). This production of metal sulfide was generally comparable to that of HgS in our previous studies [13–15], and it was produced to a higher level in the more rapidly growing eukaryotic cell treatments (Figure 2A & B). The cyanobacterium, Synechococcus, was able to synthesize significantly higher amounts of metal sulfide over time under all investigated conditions, although it is much less tolerant to Cd(II) than the eukaryotic species. Heavy metals are known to bind with low molecular weight thiol compounds such as glutathione and phytochelatins [28, 29]. The latter are low molecular weight metallothioneins synthesized from glutathione . Like metal sulfides, per se, metals bound in this way are more stable and less likely to cause oxidative damage. Cytosolic fractions taken from species of cyanobacteria and algae after exposure to Cd(II) have shown that approximately 30% of these metals are bound with metallothioneins, including phytochelatins [30–32]. Metallothioneins can exist as low and high molecular weight variants. In low molecular weight forms the metal is bound to thiol groups, whereas in the high molecular weight forms, additional inorganic sulfur is incorporated into the complexes  which appear to stabilize and improve detoxification. Interestingly, it is this pool of inorganic sulfur that is probably associated with Cd to form CdS.
Sulfate treatments resulted in more production of cellular metal sulfide by comparison with the treatment with Cd(II) alone in all investigated species (Figure 2). This took longer to become apparent in the cyanobacterial species (48 h, Figure 2C) where significant differences from the control also occurred in the sulfite and cysteine treatments. The latter was not the case for Chlamydomonas or Cyanidioschyzon. Here again, this could be accounted for by sulfur metabolism differences between cyanobacteria and algae, or possibly distinct tolerances to the toxic effects of these metabolites. High rates of sulfite assimilation into amino acids  and high expression of SSU1, a sulfite efflux gene , are known to result in lower toxicity to sulfite in yeast. Similar mechanisms may also occur in Synechococcus.
The thermophilic red microalga, Cyanidioschyzon, was capable of biotransforming approximately three times as much Cd(II) into metal sulfide as the mesophilic green alga, Chlamydomonas, when both were grown in 100 μM Cd(II). This ability may be accounted for by its adaptation to sulfur-rich hot springs . In fact, the Cyanidium medium  used to grow Cyanidioschyzon contains over an order of magnitude more sulfate than the high salt medium conventionally used for Chlamydomonas. The sensitivity of Synechococcus to Cd(II) is much higher than in the eukaryotic species. Nevertheless, metal biotransformation into sulfide by this species was only about half of that for Chlamydomonas, indicating that although sensitive to cadmium, it was able to transform a high proportion of the Cd(II) into metal sulfide. The fact that Synechococcus can convert a relatively high amount of Cd(II) into metal sulfide while remaining very sensitive to Cd(II), might be attributed to a relatively high susceptibility to displacement of metals by Cd as cofactors in photosynthetic and other metabolic enzymes, and to disruption of membrane function . Similarly, this could account for the differences between the algal species.
The first report of acid labile sulfide in living organisms was in association with metallothioneins and phytochelatins in fission yeast , and it is known that metallothionein gene amplification can confer resistance to cadmium in Synechococcus PCC 6301 . Algal phytochelatins bind cadmium in relatively low metal to peptide amounts  and it is likely that CdS formed in the organisms in the present study are mainly in the form of precipitated nanoparticles, examples of which have been reported in as diverse organisms as Klebsiella, marine microalgae , tomatoes  and mustard plants . This, however, remains to be confirmed.
Most organisms absorb sulfur from the environment in the form of inorganic sulfate and active transport systems for sulfate uptake have been investigated extensively in algae [44–46], bacteria , yeast , and higher plants [49, 50].
Algae and cyanobacteria appear to undergo sulfur assimilation in a similar manner [51, 52]. Absorbed sulfate is first converted to adenylylsulfate (APS) by ATP sulfurylase. Adenylylsulfate is then further reduced by APS reductase to yield sulfite which in turn is converted to sulfide by sulfite reductase. This sulfide is immediately transferred to the serine acetyltransferase/O-acetylserine(thiol)lyase bi-enzymatic complex (SAT-OASTL) that covalently binds it to serine to produce cysteine [50, 51]. Because all assimilated sulfate is converted into cysteine via SAT-OASTL, measuring these enzymes’ coupled activity provides a convenient means of comparing sulfate assimilation between species in response to various treatments.
The activities of SAT-OASTL in Chlamydomonas were similar to those of Ravina and colleagues  in the non-metal controls. In addition, their sulfite treatment had a similar activity to the pre- and simultaneously fed sulfite treatment in the present study. However, it is difficult to assess the effect of sulfite on specific enzymes because of its cellular toxicity (Figure 1A), something that was not considered in the previous study. The highest enzyme activities occurred when Cd(II) was provided without any supplemental sulfur containing compounds, a state in which sulfur reserves of the cells would be consumed in the CdS synthetic process (Figure 2A). Sulfur starvation has been previously shown to significantly up-regulate OASTL activity  as has Cd(II) exposure (, but this has never been studied in the context of aerobic cadmium sulfide synthesis. The highest bioconversion of Cd(II) into metal sulfide was performed when Chlamydomonas was supplemented with extra sulfate. However, this did not result in significant differences in SAT-OASTL activity from the non-metal control which was significantly lower than the Cd(II) control. This could be because Cd-elicited sulfur deprivation in the cells is compensated for by sulfate provision.
Similar to Chlamydomonas, both Cyanidioschyzon and Synechococcus possessed the highest SAT-OASTL activities during the Cd(II) control conditions. However, unlike in Chlamydomonas, simultaneous sulfate treatments had significantly higher activities than the non-metal controls (ANOVA, p < 0.05). This appears to be contradictory because these cells have higher S-nutrition than the controls and it has been shown that S-deprivation enhances OASTL activity . However, Cd-induced S-deprivation does not appear to be compensated for by the simultaneous provision of sulfate whereas extra sulfate provided by additional pre-treatments did lower enzyme activity to closer to the control levels, thereby revealing an S-nutritional effect.
Major differences occurred in the cysteine treatments between Chlamydomonas and Synechococcus that displayed expected low activities compared to controls, and the higher activities observed in Cyanidioschyzon. This could be explained by the latter species’ adaptation to sulfur-rich hot spring environments [36, 53, 54], as its SAT-OASTL activities were generally a magnitude higher than in the other species. Cyanidioschyzon enzymes need not be regulated as stringently as for Chlamydomonas and Synechococcus given that sulfur would never normally become limiting in its native environment where it could utilize the sulfur assimilation pathway for metal detoxification without experiencing the threat of sulfur depletion. Lending support to this notion is that this red alga possesses one additional SAT and two additional OASTL homologues . However, it synthesized more CdS only under sulfate-, and not sulfite- or cysteine-supplemented conditions in a similar manner to Chlamydomonas, and in contrast to Synechococcus where all conditions gave significant increases in acid labile sulfide production (Figure 2C).
The extracted activity of SAT-OASTL indicates that these enzymes do play a role in the production of required assimilated sulfur for cadmium sulfide as it was highest when cells were exposed to Cd(II) without sulfate supplementation. Higher plants that have been genetically engineered to have higher levels of these enzymes have shown some increased resistance to Cd(II) [11, 56] and other metals . Bearing in mind that in vivo activity would be distinct from extracted activity because of, among other things, different substrate concentrations, it is likely that sulfur flux through SAT-OASTL would be higher in the sulfate supplemented cells, which could contribute to the respective elevated CdS production.
Enzymatic sulfide production
Hydrogen sulfide, traditionally considered a toxic compound, has recently been implicated in cellular signaling [58, 59]. However, it would be expected that metal sulfide biosynthesis should require more sulfide than signaling processes. Several metabolic sources of sulfide have been proposed  and of these, cysteine desulfhydrase activity is the most evident and is accentuated by feeding with cysteine . In addition, there is some evidence that sulfide is released during excess sulfate nutrition which can be through provision of sulfate or sulfur dioxide/sulfite . This appears to be because of inadequate cellular supplies of O-acetylserine  and as such, OASTL cannot utilize all the H2S generated by sulfite reductase. This could have occurred in the sulfate supplemented cultures of this study, particularly in the case of Cyanidioschyzon where the sulfate concentration was 108.6 mM, however sulfate in the media of the other two species was relatively low. Other metabolic sources of significant amounts of H2S are speculative.
The assayed activity of cysteine desulfhydrase was generally much higher in Cyanidioschyzon than in Chlamydomonas and Synechococcus (Figure 4) as was the case for SAT-OASTL (Figure 3). This further indicates adaptation to high sulfur environments that accounts for its elevated ability to biotransform Cd(II) into metal sulfide which is insoluble and therefore, non-toxic. The highest activity of all was at only 6 h in the pre- and simultaneous treatment of Cyanidioschyzon with cysteine, but this did not result in an enhanced production of metal sulfide. In fact, both types of cysteine treatments in all species had relatively high cysteine desulfhydrase activities at 6 h with no enhanced metal sulfide production. Unfortunately, treatments with lower amounts of cysteine did not result in detectable increases in metal sulfide production (data not shown). This implies that the enzyme may not be involved in the supply of sulfide for CdS synthesis, or that excess cysteine is inhibitory. The latter is likely because supplementation with sulfate prior to and during Cd(II) exposure resulted in the highest desulfhydrase activities after 24 h in all three species as well as the highest production scenarios for metal sulfide. In addition, the simultaneous addition of extra sulfate with Cd(II) also resulted in relatively high extracted enzyme activity. This is consistent with the fact that Escherichia coli genetically engineered to contain unregulated cysteine desulfhydrase do produce elevated amounts of CdS [64, 65], and the formation of CdS nanoparticles appears to increase with extractable cysteine desulfhydrase activity in the photosynthetic bacterium Rhodopseudomonas palustris. Although the accumulation of acid labile sulfide is high in the organisms presented in this study, it remains to be seen if they comprise CdS nanoparticles.