Overexpression of the PHO84 gene causes heavy metal accumulation and induces Ire1p-dependent unfolded protein response in Saccharomyces cerevisiae cells
- 360 Downloads
Pho84p, the protein responsible for the high-affinity uptake and transport of inorganic phosphate across the plasma membrane, is also involved in the low-affinity uptake of heavy metals in the Saccharomyces cerevisiae cells. In the present study, the effect of PHO84 overexpression upon the heavy metal accumulation by yeast cells was investigated. As PHO84 overexpression triggered the Ire1p-dependent unfolded protein response, abundant plasma membrane Pho84p could be achieved only in ire1Δ cells. Under environmental surplus, PHO84 overexpression augmented the metal accumulation by the wild type, accumulation that was exacerbated by the IRE1 deletion. The pmr1Δ cells, lacking the gene that encodes the P-type ATPase ion pump that transports Ca2+ and Mn2+ into the Golgi, hyperaccumulated Mn2+ even from normal medium when overexpressing PHO84, a phenotype which is rather restricted to metal-hyperaccumulating plants.
KeywordsPHO84 Overexpression Heavy metal accumulation Saccharomyces cerevisiae
Transition metals such as Mn, Cu, Co, Zn, and Fe are required in trace amounts for normal metabolism, having important biological roles in stabilizing the structure of biomolecules, or acting in the catalytic center of numerous enzymes. Other heavy metals, such as Cd, Hg, and Pb, are not essential for life, but they can compete with the essential trace elements either for the transport systems or in binding to various biomolecules. Essential or not, when present in the environment in concentrations higher than normal, the heavy metals become toxic, causing serious damages to organisms due to nonspecific binding to proteins or by interfering to other metals’ metabolism.
The budding yeast Saccharomyces cerevisiae has been extensively used as a model organism to study the molecular mechanisms involved in the heavy metal transport (Van Ho et al. 2002; Feldmann 2005) and homeostasis (Ratherford and Bird 2004; Massé and Arguin 2005; Eide 2009; Reddi et al. 2009; Rosenfeld et al. 2010). In this context, the high-affinity inorganic phosphate transporter encoded by the PHO84 gene (Bun-ya et al. 1991) was found to act also as a low affinity transporter for Mn2+, Cu2+, Co2+, and Zn2+, playing a role in Mn2+ homeostasis in the S. cerevisiae cells, predominantly under metal surplus conditions (Jensen et al. 2003). Thus, yeast cells devoid of the Pho84p are more resistant to Mn2+, Cu2+, Co2+, and Zn2+ as the PHO84 gene disruption results in reduced metal-ion accumulation (Jensen et al. 2003).
Due to their potential harmful effect, the essential elements are not accumulated by the yeast cells, hence their low potential to act as heavy metal scavengers or bioremediators (Ruta et al. 2010), in spite of their high biosorptive capacity (Machado et al. 2008; 2009). Pmr1p, the P-type ATPase ion pump responsible for transporting Ca2+ and Mn2+ into the Golgi (Rudolph et al. 1989; Antebi and Fink 1992; Sorin et al. 1997; Mandal et al. 2000), is one of the key elements in the heavy metal extrusion from the cell. Pmr1p provides a major route for cellular detoxification of Mn2+ (and to a lesser extent of Cd2+, Co2+, and Cu2+) by transporting the excess cytosolic ions into the Golgi, from where they exit the cell via the secretory pathway vesicles (Lapinskas et al. 1995; Dürr et al. 1998; Mandal et al. 2000; Culotta et al. 2005; Lauer Júnior et al. 2008).
Although the link between inorganic phosphate and metal-ion transport across the plasma membrane has long been considered (Borst-Pauwels 1981; Kihn et al. 1988) and there is evidence that phosphate accumulation affects heavy metal homeostasis in the yeast cells (Rosenfeld et al. 2010), a correlation between the PHO84 overexpression and the heavy metal accumulation has not been reported. Pho84p is a transmembrane protein that is processed in the ER before being targeted to the plasma membrane (Bun-ya et al. 1996; Miller et al. 2005). Overproduction of secretory or plasma membrane proteins often overwhelms the ER processing capacity, resulting in misfolded proteins readily sensed by the Ire1p, the key component of the unfolded protein response (UPR) pathway (Kohno 2010 and references within). In yeast, activated Ire1p transmits the signal by removing an intron from the HAC1 mRNA that translates into the Hac1p transcription activator of the UPR target genes (Cocs and Walter 1996; Kawahara et al. 1997), many of which possess an UPR element (UPRE) in their promoter sequence (Mori et al. 1992).
In this study, the effect of PHO84 overexpression on heavy metal accumulation by yeast cells was investigated, but as the PHO84 overexpression elicited an Ire1p-mediated UPR response, abundant plasma membrane Pho84p could be achieved only in ire1Δ knockout cells. We show here that under metal surplus, the yeast cells overexpressing the PHO84 gene acquire the capacity to accumulate Mn2+, Cu2+, or Co2+, capacity which is augmented in the ire1Δ cells, while under normal metal concentration the pmr1Δ cells become Mn2+ hyperaccumulators.
Materials and methods
Strains and culture conditions
The S. cerevisiae strains used in this study were isogenic with the “wild-type” (WT) parental strain BY4741 (MATa, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0; Brachmann et al. 1998). The knockout mutant strains used were Y06524 (BY4741, pho84::kanMX4), Y04534 (BY4741, pmr1::kanMX4), Y01907 (BY4741, ire1::kanMX4). All strains were obtained from EUROSCARF (European S. cerevisiae Archive for Functional Analysis; Institute of Molecular Biosciences Johann Wolfgang Goethe-University Frankfurt, Germany). Cell storage, growth, and manipulation were done as described (Sherman et al. 1986). Strains were grown in standard yeast extract–polypeptone–dextrose (YPD), synthetic dextrose (SD), or synthetic galactose (SG) media supplemented with the necessary amino acids. For solid media, 2% agar was used. For growth improvement, the synthetic media were supplemented with 0.1% yeast extract.
The PHO84 open reading frame was amplified by PCR from genomic DNA using primers PHO84rec1sens (5′GAATTCGATATCAAGCTTATCGATACCGTCGACAATGAGTTCCGTCAATAAAGA3′) and PHO84rec2AS (5′GCGTGACATAACTAATTACATGACTCGAGGTCGACTTATGCTTCATGTTGAAGTT3′), the underlined sequence being homologous to the target vectors pGREG505 (LEU2 marker) and pGREG506 (URA3 marker). For cloning into pGREG600 (URA3 marker), to construct the PHO84-GFP fusion, PHO84rec1sens and the reverse primer PHO84rec4AS 5′GTTCTTCTCCTTTACTCATTCTCGAGGTCGACTGCTTCATGTTGAAGTTGAGAT3′ were used (Jansen et al. 2005). The PCR fragments were purified by Wizard SV gel and PCR Clean-up System (Promega) and introduced into pGREG505, pGREG506, or pGREG600 by cotransformation into yeast cells (Jansen et al. 2005), thus generating pPHO84-505, pPHO84-506, and pPHO84-600, respectively. The plasmids were rescued from yeast cells and subsequently amplified and purified from E. coli cells. The pGREG plasmid series was purchased from EUROSCARF. Plasmid pCZY1 (URA3, 2 μ) harboring the lacZ reporter gene driven by the CYC core promoter fused with the UPRE was generously provided by Professor Kenji Kohno (Nara Institute of Science and Technology, Japan) and used to monitor cellular UPR activity (Mori et al. 1992). Yeast transformation was performed by a modified lithium acetate method (Schiestl and Gietz 1989), and transformants were selected for growth on SD medium lacking the appropriate selective nutrients.
Overnight precultures were inoculated in fresh media at density 106 cells/mL, then cells were incubated with shaking (200 rpm) at 28°C for 2 h before being shifted to selective SG to allow gene induction. The influence of metals on cell growth in liquid media was monitored at time intervals by determining the optical density of cellular suspension at 660 nm (Shimadzu UV-VIS spectrophotometer, UV mini 1240). For spot assay, fresh cell cultures of OD660∼1 were diluted 10-, 100-, 1,000-, and 10,000-fold and stamped on agar plates using a replicator. For viability test, samples (200 μL; three replicates) were taken at defined intervals of time, serially diluted with sterile deionized water and plated on YPD/agar. After 3–4 days of incubation at 28°C, the colonies were counted. Original cell suspensions had viability higher than 99%.
Metal accumulation by living cells
Overnight precultures were diluted in fresh SD selective medium to 106 cells/mL. The cells were subsequently incubated with shaking for 2 h at 28°C before being washed and shifted to SG selective medium, used to induce gene expression via GAL1 promoter. Six hours after the galactose shift, metal ions (in the chloride forms) were added from sterile stocks. Cells grown in media containing various concentrations of MnCl2, CoCl2, or CuCl2 were harvested by centrifugation and were washed three times with 10 mM 2-(N-morpholino)ethanesulfonic acid (MES)–Tris buffer, pH 6.0, at 0–2°C. All centrifugations (1 min, 5,000 rpm) were done at 4°C. Cells were finally suspended in 10 mM MES–Tris buffer, pH 6.0 (109 cells/mL) and used for metal and cell protein assay. Metal analysis of whole cell was done using an instrument with a single collector, quadrupole inductively coupled plasma with mass spectrometry (ICP-MS; Perkin-Elmer ELAN DRC-e) with axial field technology for trace elements, rare earth elements, and isotopic analyses. The analyses were performed after digestion of cells with 65% ultrapure HNO3 (Merck). Standard solutions were prepared by diluting a 10-μg/mL multielement solution (Multielement ICP Calibration Standard 3, matrix 5% NNO3, Perkin Elmer Pure Plus). The metal cellular content was normalized to total cellular proteins, which were assayed as described by Bradford (1976). Values are expressed as the mean ± standard deviation of duplicate determinations of three independent yeast transformants. The experiments were repeated on three different days, with similar results.
For the detection of Pho84-GFP fluorescence, cells transformed with pPHO84-600 were grown overnight in SD-Ura, diluted to 106 cells/mL and grown for 2 h before being shifted to SD-Gal. Cells were examined with a Leica fluorescent microscope system (Leica DM 1000; Wetzlar, Germany) equipped with an HBO-100 mercury lamp and a Leica DFC 295 camera. To detect the GFP signals, a GFP filter set (excitation filter BP 470/40, dichromatic mirror 500, and suppression filter BP 525/50) was used. The microscopic photographs were processed using the Leica Application Suite software.
Phosphate removal test
Phosphate removal by cells from the incubation medium was done as described by Watanabe et al. (2008), with slight modifications. Yeast strains were precultivated in SD-Ura for 12 h. The cells were harvested, washed twice with deionized water, and then used to inoculate (OD660 1.0) 5 mL SG-Ura prepared in 0.1 mM phosphate buffer, pH 5.5. The cells were grown at 28°C, with shaking (200 rpm), and aliquots were taken at 4 and 6 h from the galactose shift. The cells were removed by centrifugation, and the remnant phosphate was determined as described (Nyren et al. 1986). Values are expressed as the mean ± standard deviation of duplicate determinations of three independent yeast transformants. The experiments were repeated on three different days, with similar results.
The β-galactosidase activity of yeast extract was carried out on permeabilized cells as described (Amberg et al. 2005). Values are expressed as the mean ± standard deviation of duplicate determinations of three independent yeast transformants.
Reproducibility of the results
All experiments were repeated, independently, on three different days. For each individual determination, values were expressed as the mean ± standard deviation of duplicate measurements on three independent yeast transformants (n = 6). The data reported for viability are the mean ± SD of colony numbers obtained in triplicate from three independent transformants (n = 9). The observed trends were fully consistent among the experiments performed on different days, and a representative example is shown.
Multiple comparisons were performed with Student’s t test. The differences were considered to be significant when p < 0.05. Data analysis was performed with Statistical Package for Social Science 15.0 (SPSS 15.0) for Windows.
Overexpression of PHO84 from a GAL1 promoter augments the heavy metal accumulation by yeast cells
PHO84 overexpression triggers the UPR pathway
Under the conditions described in Fig. 1c, the metal accumulation by the cells overexpressing PHO84 was constantly but moderately higher than that observed for WT cells. When the tests were repeated on different days, no obvious day-to-day variations were observed. Statistical analysis revealed that PHO84 overexpression enhanced the Mn2+ accumulation (2.8 ± 0.31) times, the Co2+ accumulation (2.6 ± 0.28) times, and the Cu2+ accumulation (1.7 ± 0.24) times (p < 0.05). Pho84p is a transmembrane protein processed in the ER before being targeted to the plasma membrane (Bun-ya et al. 1996; Miller et al. 2005), therefore Pho84p overproduction may overwhelm the ER processing capacity resulting in protein misfolding and UPR onset. To test this possibility, the plasmid containing the GAL1-PHO84 fusion was co-expressed with a plasmid harboring the reporter LacZ gene under the control of CYC core promoter fused with the UPRE. This latter system is suitable for quantification of Ire1p-mediated UPR by measuring the β-galactosidase activity of cells that express it (Mori et al. 1992). We found that PHO84 overexpression from the GAL1 promoter activated the UPR in wild-type cells (WTUR/PHO84-OE), as demonstrated by the high values of β-galactosidase of the permeabilized cells. This activation was Ire1p dependent, since the β-galactosidase activity in the ire1Δ cells overexpressing PHO84 (ire1ΔUR/PHO84-OE) was very low (Fig. 2a). To check the Pho84p localization in PHO84 overexpressing cells, we constructed a PHO84-GFP fusion driven by the GAL1 promoter, and cells expressing this fusion were visualized by fluorescent microscopy. It was noticed that the WT cells overexpressing the PHO84-GFP fusion exhibited a fluorescent halo, depicting the plasma membrane localization of the chimeric protein. Nevertheless, the fluorescence was faint compared to that observed for the ire1Δ cells overexpressing the PHO84-GFP fusion construct (Fig. 2b). This observation suggested that overproduction of Pho84p is controlled in the wild-type cells via Ire1p activation of the UPR and that high Pho84p at plasma level can be achieved only in cells defective in the UPR.
Overexpressing PHO84 in ire1Δ cells results in enhanced metal accumulation
The pmr1Δ cells overexpressing PHO84 are Mn2+ hyperaccumulators
Metal accumulation by cells grown in standard SG-Ura medium
Accumulated metal (nmol/mg total protein)
4.89 ± 0.04
5.2 ± 0.06
3.2 ± 0.12
4.1 ± 0.28
4.05 ± 0.12
4.89 ± 0.89
3.11 ± 0.01
2.12 ± 0.01
2.1 ± 0.056
1.9 ± 0.12
2.19 ± 0.023
1.9 ± 0.34
12.2 ± 0.35
16.8 ± 0.46
6.8 ± 0.32
8.24 ± 0.44
5.1 ± 0.12
6.2 ± 0.34
36.8 ± 1.02
40.23 ± 1.14
12.1 ± 0.34
14.12 ± 0.56
8.2 ± 0.23
10.87 ± 1.16
16.97 ± 1.24
24.54 ± 1.67
7.38 ± 1.87
8.8 ± 1.21
6.4 ± 0.28
8.86 ± 0.87
46.22 ± 0.78
56.8 ± 2.12
30.9 ± 0.72
33.12 ± 0.84
24.9 ± 0.52
33.2 ± 2.43
1797.88 ± 3.9
1976.2 ± 4.23
8.22 ± 0.28
10.28 ± 1.12
17.7 ± 0.32
22.22 ± 1.86
In this work it was shown that PHO84 overexpression accompanied by increased targeting of Pho84p to the plasma membrane augmented the heavy metal accumulation by the yeast cells. Obtaining heavy metal-accumulating mutants is a prime objective for technologies dealing with bioremediation of contaminated sites (metal excess) or with bioextraction from depleted sites (metal deficiency). Nevertheless, bioaccumulation has bioremediation significance only when correlated with increased (gained) tolerance to the accumulated substance. Similarly to heavy metal-accumulating plants (Baker 2002; Baker and Brooks 1989), both wild-type and UPR-disrupted (ire1Δ) cells overexpressing PHO84 could accumulate Mn2+, Co2+, or Cu2+ without developing toxicity symptoms (Figs. 1 and 3). The metal accumulation is probably facilitated by the fact that the ions are taken up by the cell in a bound form, as suggested by the possibility of a metal phosphate co-transport, which limits the entry of the osmotically free metal ions (the more toxic form) into the cytoplasm. Alternatively, as Pho84p is primarily involved in the high-affinity phosphate transport, it is also possible that the metal ions penetrate the cells concomitantly with the phosphate, the subsequent interaction taking place in the cytosol. The solubility of Mn2+, Co2+, or Cu2+ phosphates is low, limiting the occurrence of osmotically free metal ions (the more toxic form) in the cytoplasm, the place where many metabolically active enzymes or other biomolecules reside. The role of (poly)phosphate in alleviating metal toxicity is well documented (Borst-Pauwels 1981; Kihn et al. 1988; Farcasanu et al. 1996; Rosenfeld et al. 2010) and developing a system that co-transports phosphate with high affinity and metal ions with low affinity can be considered as a second protective action of a transporter whose main function is nutrient transport under limiting conditions. It was interesting to see that PHO84 overexpression significantly augmented the metal accumulation by ire1Δ cells under metal surplus conditions. Under normal conditions though, IRE1 deletion resulted in significantly higher manganese accumulation than simply overexpressing PHO84 in wild-type cells (∼10-fold compared to ∼4-fold, respectively, Table 1), suggesting that Ire1p-dependent UPR may clear the endogenous Pho84p. It was also striking that in standard medium, overexpression of PHO84 in ire1Δ cells does not result in significantly enhanced Mn levels relatively to the ire1Δ cells expressing the empty vector (Table 1). This observation suggested the possibility that endogenous Pho84p is already overexpressed in ire1Δ cells, thus rendering them potential candidates as bioextractors of manganese from low concentration environments. Nevertheless, this does not mean that there is no need to overexpress PHO84 in ire1Δ cells to obtain accumulating mutants. As Pho84p exhibits only low affinity for divalent cations (Jensen et al. 2003), the overexpression from a strong promoter such as GAL1 promoter may simply be superfluous for cation uptake from low concentration environments. The advantage of overexpressing PHO84 in ire1Δ cells can be seen when it comes to higher concentrations: under such circumstances, the low-affinity transporter seem to have the right abundance to engulf the surplus cations.
The pmr1Δ cells overexpressing PHO84 deserve special attention, since under normal growth conditions they tend to hyperaccumulate Mn2+, i.e., over 300 times more than the wild-type cells (Table 1), a value that corresponds to the definition of a hyperaccumulator given for plants (Baker and Brooks 1989). Considering the haploid yeast cell size and protein composition (Sherman 2002), the average Mn2+ concentration reached within the cell in conditions described in Table 1 would be grossly a staggering 150 mM, which is 20,000 more than the 0.75 μM detected in the SG-Ura medium used. The hyperaccumulating activity of pmr1Δ cells seemed to be restricted to Mn2+, and this was not surprising, considering the Pmr1p preference for Mn2+ (Lapinskas et al. 1995; Dürr et al. 1998). When this pump is lacking, the Mn2+ imported via the more-abundant-than-usual plasma membrane Pho84p cannot be extruded, but instead accumulates in the cell. Even though Mn2+ hyperaccumulation was accompanied by slow growth and loss of viability (thus enlarging the selected group of kamikaze cells, Ruta et al. 2010), controlling the level of PHO84 expression may improve cell growth, while maintaining the hyperaccumulating capacity. Apart from few marine organisms (Gifford et al. 2007), heavy metal hyperaccumulation is a phenotype characteristic to plants, and extrapolating it to a friendly microorganism like S. cerevisiae would open new possibilities for many fields, including biotechnology.
Manganese is very common in the environment, and much used in the industry or as fuel additive. It is an essential nutrient but also has the potential to produce neurotoxic effects when it accumulates in an organism, especially in the brain (for review, Takeda 2003). There are few plants that hyperaccumulate Mn from both contaminated soils (1% Mn) or from soils with normal concentration (Reeves 2006). Such plants are predominantly woody (Fernando et al. 2008, 2009) and hence unsuited to short-term controlled study. The use of friendly microorganisms as an alternative for hyperaccumulating plants is therefore an attractive possibility, and engineered yeast cells may be the organisms of choice. Nevertheless, the use of genetically modified cells would restrict the applicability of the method, unless strictly controlled fermenters are to be used, making this cellular system best suited for bioremediation of contaminated waters. The European and US regulations recommend a limit of 0.05 mg/L manganese (approximately 1 μM, similar to the concentration in yeast growth medium) in the drinking water; but for many industrial purposes, the manganese content should not exceed 0.01 to 0.02 mg/L, and in some cases, even this is considered excessive. In this respect, two PHO84-overexpressing yeast strains described in this study seem to be of relevance for potential usage in decreasing the manganese concentrations of contaminated waters.
The authors are grateful to Professor Kenji Kohno (Nara Institute of Science and Technology, Japan) for generously providing plasmid pCZY1. This work was supported by the Ministry of Education and Research of Romania through the grant-in-aid PNII Idei_965, 176/2007 and by the postdoctoral program POSDRU/89/1.5/S/60746, from the European Social Fund.
- Amberg DC, Burke DJ, Strathern JN (2005) Methods in yeast genetics. A Cold Spring Harbor Laboratory course manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, pp 135–139Google Scholar
- Antebi A, Fink GR (1992) The yeast Ca2+-ATPase homologue, PMR1, is required for normal Golgi function and localizes in a novel Golgi-like distribution. Mol Biol Cell 3:633–654Google Scholar
- Baker AJM, Brooks RR (1989) Terrestrial higher plants which hyperaccumulate metallic elements—a review of their distribution, ecology and phytochemistry. Biorecovery 1:81–126Google Scholar
- Baker AJM (2002) The use of tolerant plants and hyperaccumulators. In: Wong MH, Bradshaw AD (eds) Restoration and management of derelict land: modern approaches. World Scientific Publishing Co., Singapore, pp 138–148Google Scholar
- Borst-Pauwels GW (1981) Ion transport in yeast. Biochim Biophys Acta 650:88–127Google Scholar
- Bun-ya M, Nishimura M, Harashima S, Oshima Y (1991) The PHO84 gene of Saccharomyces cerevisiae encodes an inorganic phosphate transporter. Mol Cell Biol 11:3229–3238Google Scholar
- Bun-ya M, Shikata K, Nakade S, Yompakdee C, Harashima S, Oshima Y (1996) Two new genes, PHO86 and PHO87, involved in inorganic phosphate uptake in Saccharomyces cerevisiae. Curr Genet 29:344–351Google Scholar
- Dürr G, Strayle J, Plemper R, Elbs S, Klee SK, Catty P, Wolf DH, Rudolph HK (1998) The medial-Golgi ion pump Pmr1 supplies the yeast secretory pathway with Ca2+ and Mn2+ required for glycosylation, sorting, and endoplasmic reticulum-associated protein degradation. Mol Biol Cell 9:1149–1162Google Scholar
- Feldmann H (2005) Yeast transport. In: Yeast molecular biology, a short compendium on basic features and novel aspects. Adolf-Butenandt-Institute, University of Munich, pp 10–17. http://biochemie.web.med.uni-muenchen.de/Yeast_Biol/
- Fernando DR, Woodrow IE, Jaffré T, Dumontet V, Marshall AT, Baker AJM (2008) Foliar manganese accumulation by Maytenus founieri (Celastraceae) in its native New Caledonian habitats: populational variation and localization by X-ray microanalysis. New Phytol 177:178–185Google Scholar
- Guthrie C, Fink GR (eds) (1991) Guide to yeast genetics and molecular biology. Academic, New YorkGoogle Scholar
- Kawahara T, Yanagi H, Yura T, Mori K (1997) Endoplasmic reticulum stress-induced mRNA splicing permits synthesis of transcription factor Hac1p/Ern4p that activates the unfolded protein response. Mol Biol Cell 8:1845–1862Google Scholar
- Lapinskas PJ, Cunningham KW, Liu XF, Fink GR, Culotta VC (1995) Mutations in PMR1 suppress oxidative damage in yeast lacking superoxide dismutase. Mol Cell Biol 15:1382–1388Google Scholar
- Machado MD, Janssens S, Soares HM, Soares EV (2009) Removal of heavy metals using a brewar’s yeast strain of Saccharomyces cerevisiae: advantages of using dead biomass. J Appl Biotech 106:1792–1804Google Scholar
- Mori K, Sant A, Kohno K, Normington K, Gething MJ, Sambrook JF (1992) A 22 bp cis-acting element is necessary and sufficient for the induction of the yeast KAR2 (BiP) gene by unfolded proteins. EMBO J 11:2583–2593Google Scholar
- Sherman F, Fink GR, Hicks JB (1986) Methods in yeast genetics. Cold Spring Harbor Laboratory Press, Cold Spring HarborGoogle Scholar