DOI: 10.1126/science.287.5450.122. A Short Fe-Fe Distance in Peroxodiferric Ferritin: Control of Fe Substrate Versus Cofactor Decay?

A Short Fe-Fe Distance in Peroxodiferric Ferritin: Control of Fe Substrate Versus Cofactor Decay?

10.1126/science.287.5450.122

Crossref journal-article

American Association for the Advancement of Science (AAAS)

Science (221)

Abstract

The reaction of oxygen with protein diiron sites is important in bioorganic syntheses and biomineralization. An unusually short Fe-Fe distance of 2.53 angstroms was found in the diiron (μ-1,2 peroxodiferric) intermediate that forms in the early steps of ferritin biomineralization. This distance suggests the presence of a unique triply bridged structure. The Fe-Fe distances in the μ-1,2 peroxodiferric complexes that were characterized previously are much longer (3.1 to 4.0 angstroms). The 2.53 angstrom Fe-Fe distance requires a small Fe-O-O angle (∼106° to 107°). This geometry should favor decay of the peroxodiferric complex by the release of H 2 O 2 and μ-oxo or μ-hydroxo diferric biomineral precursors rather than by oxidation of the organic substrate. Geometrical differences may thus explain how diiron sites can function either as a substrate (in ferritin biomineralization) or as a cofactor (in O 2 activation).

Bibliography

Hwang, J., Krebs, C., Huynh, B. H., Edmondson, D. E., Theil, E. C., & Penner-Hahn, J. E. (2000). A Short Fe-Fe Distance in Peroxodiferric Ferritin: Control of Fe Substrate Versus Cofactor Decay? Science, 287(5450), 122â125.

Authors 6

Jungwon Hwang (first)
Carsten Krebs (additional)
Boi Hanh Huynh (additional)
Dale E. Edmondson (additional)
Elizabeth C. Theil (additional)
James E. Penner-Hahn (additional)

References 54 Referenced 163

10.1021/cr9500489
10.1021/cr00027a011
; G. S. Waldo and E. C. Theil in Bioinorganic Systems vol. 5 of Comprehensive Supramolecular Chemistry K. S. Suslick Ed. (Pergamon Oxford 1996) pp. 65–89.
Animals produce H- L- and M-type ferritin subunits; the H and M types have a functional diiron site. The subunits coassemble to form cell-specific mixtures. Although the functional significance of the subunit ratio is not fully understood this ratio is controlled precisely and is fixed at the time of cell differentiation. Plants and bacteria appear to have only a single ferritin subunit of the H type which contains a functional diiron site.
10.1021/bi980847w
10.1021/bi990095l
10.1021/ja00146a002
10.1021/ja9839522
10.1021/ja973651c
10.1021/bi981838q
10.1021/bi981839i
10.1021/bi9914199
10.1021/ja9718230
; P. J. Riggs-Gelasco thesis University of Michigan (1994).
An O 2 -saturated apoferritin solution in 0.2 M Mops buffer (pH 7.0) and 0.2 M NaCl was mixed at 23°C with an equal volume of O 2 -saturated 57 Fe-enriched (95% enrichment) FeSO 4 in 0.2 M NaCl and 3.5 mM H 2 SO 4 . The final protein concentration after mixing was about 49 μM of 24-polymer ferritin molecule. The Fe:ferritin molecule ratio was kept at 36:1. The reaction of apoferritin with Fe 2+ and O 2 was quenched at 25 ms when the peroxodiferric intermediate had accumulated maximally and at 1 s when the intermediate had decayed.
The experiment was conducted at 3.0 GeV and 50 to 100 mA with the use of a Si(220) double-crystal monochromator that was detuned 50% with an Fe foil internal calibration (first inflection point at 7111.2 eV). We used a 13-element Ge solid-state fluorescence detector (incident count rate <60 kHz per channel about 3-kHz Fe Kα fluorescence counts) at 10 K. The step size was 10 eV in the pre-edge region (6880 to 7080 eV) 0.35 eV in the edge region (7080 to 7140 eV) and 0.05 Å −1 in the EXAFS region (2 to 13 Å −1 ). Integration times ranged from 1 s in the pre-edge region to 20 s at k = 13 Å −1 where k is the photoelectron wavevector.
XANES data were normalized to tabulated x-ray cross sections EXAFS data were extracted by subtracting a first-order pre-edge polynomial and then a three-region cubic spline was fitted to the EXAFS. Data converted to k space with E 0 = 7130 eV and weighted by k 3 . k space data (1.5 to 11.9 12.4 and 12.7 Å −1 for −O 2 +O 2 25 ms and +O 2 1 s samples respectively) were Fourier transformed to give the R space data in Fig. 2.
Bond lengths coordination numbers and Debye-Waller factors determined from fitting Fourier-filtered data ( R = 1.0 to 3.2 1.0 to 3.3 and 1.0 to 3.6 Å for the −O 2 +O 2 25 ms and +O 2 1 s samples respectively) with the use of theoretical amplitude and phase functions calculated with FEFF v.6.01. [
10.1021/ja00014a001
] with the scale factor (0.9) and ΔE 0 (10 eV) calibrated from fits to data for crystallographically characterized models. Coordination numbers were fixed at reasonable values and only R and σ were treated as freely variable parameters. Identical metric parameters were obtained for fits to filtered and unfiltered data. A weighted goodness of fit F ′ = F 2 /ν = F 2 /( N idp − N var ) was used to evaluate the quality of the fit [
10.1016/0010-8545(95)01144-E
] where F is the root mean square difference between the fit and the data N idp is the number of independent data points and N var is the number of variable parameters.
10.1021/ic00236a025
10.1021/ja00024a031
10.1021/ja00318a021
10.1021/ja964352a
Supplemental data are available at Science Online at www.sciencemag.org/feature/data/1044838.shl.
10.1021/ic951462u
The refined Fe-O distances differ only by the estimated resolution of the data (π/2 k max ≈ 0.13 Å) which further indicates that these fits are not chemically meaningful.
10.1021/ic00071a023
The Debye-Waller factor was constrained to be the same as those found for the 25-ms sample to permit quantitative comparison of the coordination numbers.
10.1126/science.275.5299.515
10.1021/ja973115k
10.1021/ja9604370
10.1002/anie.199606181
10.1021/ja953705n
10.1021/ic00029a008
10.1021/ja00096a027
10.1021/ja00052a043
10.1021/ja00173a031
This assumes that the Fe-O-O-Fe torsion angle is zero. γ would be even smaller for nonzero torsion angles.
T. C. Brunold N. Tamura N. Kitajima Y. Moro-oka
10.1021/ja980129x
10.1021/bi00211a039
10.1021/bi973128a
Single-letter abbreviations for the amino acid residues are as follows: A Ala; D Asp; E Glu; F Phe; H His; Q Gln; S Ser; and W Trp.
10.1007/s007750050310
10.1006/jmbi.1997.0970
10.1016/S0021-9258(18)47653-3
10.1021/bi00046a028
10.1007/s007750050180
. Kinetics of diferric peroxo formation and decay in a QXXA ferritin (human H) differ from either the QXXD (frog M) or QXXS (frog H); no information is available for the QXXQ ferritin (maize). In the Escherichia coli ferritins EXXH occurs instead of QXXD QXXA or QXXS as the ligands to the second Fe and the Fe is retained at the diiron site [
10.1016/S0014-5793(98)00867-9
We thank K. Herlihy G. W. Small and A. R. Tipton for assistance in the preparation of the ferritin samples and B. G. Fox for communicating results before publication. Supported in part by grants from NIH (GM-45205 to J.E.P.-H. GM-47295 and GM-58778 to B.H.H. and D.E.E. DK-20251 to E.C.T.). X-ray absorption data were measured at Stanford Synchrotron Radiation Laboratory which is operated by the U.S. Department of Energy (DOE) Office of Basic Energy Sciences with additional support from NIH National Center for Research Resources and the DOE Office of Biological and Environmental Research.

Dates

Type	When
Created	23 years, 1 month ago (July 27, 2002, 5:40 a.m.)
Deposited	1 year, 7 months ago (Jan. 13, 2024, 4:53 a.m.)
Indexed	2 months ago (June 27, 2025, 3:24 p.m.)
Issued	25 years, 7 months ago (Jan. 7, 2000)
Published	25 years, 7 months ago (Jan. 7, 2000)
Published Print	25 years, 7 months ago (Jan. 7, 2000)

Funders 0

None

BibTeX

@article{Hwang_2000, title={A Short Fe-Fe Distance in Peroxodiferric Ferritin: Control of Fe Substrate Versus Cofactor Decay?}, volume={287}, ISSN={1095-9203}, url={http://dx.doi.org/10.1126/science.287.5450.122}, DOI={10.1126/science.287.5450.122}, number={5450}, journal={Science}, publisher={American Association for the Advancement of Science (AAAS)}, author={Hwang, Jungwon and Krebs, Carsten and Huynh, Boi Hanh and Edmondson, Dale E. and Theil, Elizabeth C. and Penner-Hahn, James E.}, year={2000}, month=jan, pages={122–125} }

JSON

{
  "indexed": {
    "date-parts": [
      [
        2025,
        6,
        27
      ]
    ],
    "date-time": "2025-06-27T19:24:40Z",
    "timestamp": 1751052280888
  },
  "reference-count": 54,
  "publisher": "American Association for the Advancement of Science (AAAS)",
  "issue": "5450",
  "content-domain": {
    "domain": [],
    "crossmark-restriction": false
  },
  "published-print": {
    "date-parts": [
      [
        2000,
        1,
        7
      ]
    ]
  },
  "abstract": "<jats:p>\n            The reaction of oxygen with protein diiron sites is important in bioorganic syntheses and biomineralization. An unusually short Fe-Fe distance of 2.53 angstroms was found in the diiron (\u03bc-1,2 peroxodiferric) intermediate that forms in the early steps of ferritin biomineralization. This distance suggests the presence of a unique triply bridged structure. The Fe-Fe distances in the \u03bc-1,2 peroxodiferric complexes that were characterized previously are much longer (3.1 to 4.0 angstroms). The 2.53 angstrom Fe-Fe distance requires a small Fe-O-O angle (\u223c106\u00b0 to 107\u00b0). This geometry should favor decay of the peroxodiferric complex by the release of H\n            <jats:sub>2</jats:sub>\n            O\n            <jats:sub>2</jats:sub>\n            and \u03bc-oxo or \u03bc-hydroxo diferric biomineral precursors rather than by oxidation of the organic substrate. Geometrical differences may thus explain how diiron sites can function either as a substrate (in ferritin biomineralization) or as a cofactor (in O\n            <jats:sub>2</jats:sub>\n            activation).\n          </jats:p>",
  "DOI": "10.1126/science.287.5450.122",
  "type": "journal-article",
  "created": {
    "date-parts": [
      [
        2002,
        7,
        27
      ]
    ],
    "date-time": "2002-07-27T09:40:21Z",
    "timestamp": 1027762821000
  },
  "page": "122-125",
  "source": "Crossref",
  "is-referenced-by-count": 163,
  "title": "A Short Fe-Fe Distance in Peroxodiferric Ferritin: Control of Fe Substrate Versus Cofactor Decay?",
  "prefix": "10.1126",
  "volume": "287",
  "author": [
    {
      "given": "Jungwon",
      "family": "Hwang",
      "sequence": "first",
      "affiliation": [
        {
          "name": "Department of Chemistry, University of Michigan, Ann Arbor, MI 48109\u20131055, USA."
        }
      ]
    },
    {
      "given": "Carsten",
      "family": "Krebs",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "Department of Physics,"
        }
      ]
    },
    {
      "given": "Boi Hanh",
      "family": "Huynh",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "Department of Physics,"
        }
      ]
    },
    {
      "given": "Dale E.",
      "family": "Edmondson",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "Departments of Biochemistry and Chemistry, Emory University, Atlanta, GA 30322, USA."
        }
      ]
    },
    {
      "given": "Elizabeth C.",
      "family": "Theil",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "Children's Hospital Oakland Research Institute, 5700 Martin Luther King Jr. Way, Oakland, CA 94609\u20131673, USA."
        }
      ]
    },
    {
      "given": "James E.",
      "family": "Penner-Hahn",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "Department of Chemistry, University of Michigan, Ann Arbor, MI 48109\u20131055, USA."
        }
      ]
    }
  ],
  "member": "221",
  "reference": [
    {
      "key": "e_1_3_1_2_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/cr9500489"
    },
    {
      "key": "e_1_3_1_2_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/cr00027a011"
    },
    {
      "key": "e_1_3_1_2_4",
      "unstructured": "; G. S. Waldo and E. C. Theil in Bioinorganic Systems vol. 5 of Comprehensive Supramolecular Chemistry K. S. Suslick Ed. (Pergamon Oxford 1996) pp. 65\u201389."
    },
    {
      "key": "e_1_3_1_3_2",
      "unstructured": "Animals produce H- L- and M-type ferritin subunits; the H and M types have a functional diiron site. The subunits coassemble to form cell-specific mixtures. Although the functional significance of the subunit ratio is not fully understood this ratio is controlled precisely and is fixed at the time of cell differentiation. Plants and bacteria appear to have only a single ferritin subunit of the H type which contains a functional diiron site."
    },
    {
      "key": "e_1_3_1_4_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi980847w"
    },
    {
      "key": "e_1_3_1_5_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi990095l"
    },
    {
      "key": "e_1_3_1_6_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja00146a002"
    },
    {
      "key": "e_1_3_1_6_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja9839522"
    },
    {
      "key": "e_1_3_1_7_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja973651c"
    },
    {
      "key": "e_1_3_1_7_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi981838q"
    },
    {
      "key": "e_1_3_1_8_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi981839i"
    },
    {
      "key": "e_1_3_1_8_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi9914199"
    },
    {
      "key": "e_1_3_1_9_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja9718230"
    },
    {
      "key": "e_1_3_1_9_3",
      "unstructured": "; P. J. Riggs-Gelasco thesis University of Michigan (1994)."
    },
    {
      "key": "e_1_3_1_10_2",
      "unstructured": "An O 2 -saturated apoferritin solution in 0.2 M Mops buffer (pH 7.0) and 0.2 M NaCl was mixed at 23\u00b0C with an equal volume of O 2 -saturated 57 Fe-enriched (95% enrichment) FeSO 4 in 0.2 M NaCl and 3.5 mM H 2 SO 4 . The final protein concentration after mixing was about 49 \u03bcM of 24-polymer ferritin molecule. The Fe:ferritin molecule ratio was kept at 36:1. The reaction of apoferritin with Fe 2+ and O 2 was quenched at 25 ms when the peroxodiferric intermediate had accumulated maximally and at 1 s when the intermediate had decayed."
    },
    {
      "key": "e_1_3_1_11_2",
      "unstructured": "The experiment was conducted at 3.0 GeV and 50 to 100 mA with the use of a Si(220) double-crystal monochromator that was detuned 50% with an Fe foil internal calibration (first inflection point at 7111.2 eV). We used a 13-element Ge solid-state fluorescence detector (incident count rate <60 kHz per channel about 3-kHz Fe K\u03b1 fluorescence counts) at 10 K. The step size was 10 eV in the pre-edge region (6880 to 7080 eV) 0.35 eV in the edge region (7080 to 7140 eV) and 0.05 \u00c5 \u22121 in the EXAFS region (2 to 13 \u00c5 \u22121 ). Integration times ranged from 1 s in the pre-edge region to 20 s at k = 13 \u00c5 \u22121 where k is the photoelectron wavevector."
    },
    {
      "key": "e_1_3_1_12_2",
      "unstructured": "XANES data were normalized to tabulated x-ray cross sections EXAFS data were extracted by subtracting a first-order pre-edge polynomial and then a three-region cubic spline was fitted to the EXAFS. Data converted to k space with E 0 = 7130 eV and weighted by k 3 . k space data (1.5 to 11.9 12.4 and 12.7 \u00c5 \u22121 for \u2212O 2 +O 2 25 ms and +O 2 1 s samples respectively) were Fourier transformed to give the R space data in Fig. 2."
    },
    {
      "key": "e_1_3_1_13_2",
      "unstructured": "Bond lengths coordination numbers and Debye-Waller factors determined from fitting Fourier-filtered data ( R = 1.0 to 3.2 1.0 to 3.3 and 1.0 to 3.6 \u00c5 for the \u2212O 2 +O 2 25 ms and +O 2 1 s samples respectively) with the use of theoretical amplitude and phase functions calculated with FEFF v.6.01. ["
    },
    {
      "key": "e_1_3_1_13_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja00014a001"
    },
    {
      "key": "e_1_3_1_13_4",
      "unstructured": "] with the scale factor (0.9) and \u0394E 0 (10 eV) calibrated from fits to data for crystallographically characterized models. Coordination numbers were fixed at reasonable values and only R and \u03c3 were treated as freely variable parameters. Identical metric parameters were obtained for fits to filtered and unfiltered data. A weighted goodness of fit F \u2032 = F 2 /\u03bd = F 2 /( N idp \u2212 N var ) was used to evaluate the quality of the fit ["
    },
    {
      "key": "e_1_3_1_13_5",
      "doi-asserted-by": "publisher",
      "DOI": "10.1016/0010-8545(95)01144-E"
    },
    {
      "key": "e_1_3_1_13_6",
      "unstructured": "] where F is the root mean square difference between the fit and the data N idp is the number of independent data points and N var is the number of variable parameters."
    },
    {
      "key": "e_1_3_1_14_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ic00236a025"
    },
    {
      "key": "e_1_3_1_14_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja00024a031"
    },
    {
      "key": "e_1_3_1_15_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja00318a021"
    },
    {
      "key": "e_1_3_1_16_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja964352a"
    },
    {
      "key": "e_1_3_1_17_2",
      "unstructured": "Supplemental data are available at Science Online at www.sciencemag.org/feature/data/1044838.shl."
    },
    {
      "key": "e_1_3_1_18_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ic951462u"
    },
    {
      "key": "e_1_3_1_19_2",
      "unstructured": "The refined Fe-O distances differ only by the estimated resolution of the data (\u03c0/2 k max \u2248 0.13 \u00c5) which further indicates that these fits are not chemically meaningful."
    },
    {
      "key": "e_1_3_1_20_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ic00071a023"
    },
    {
      "key": "e_1_3_1_21_2",
      "unstructured": "The Debye-Waller factor was constrained to be the same as those found for the 25-ms sample to permit quantitative comparison of the coordination numbers."
    },
    {
      "key": "e_1_3_1_22_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1126/science.275.5299.515"
    },
    {
      "key": "e_1_3_1_23_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja973115k"
    },
    {
      "key": "e_1_3_1_24_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja9604370"
    },
    {
      "key": "e_1_3_1_25_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1002/anie.199606181"
    },
    {
      "key": "e_1_3_1_26_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja953705n"
    },
    {
      "key": "e_1_3_1_27_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ic00029a008"
    },
    {
      "key": "e_1_3_1_28_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja00096a027"
    },
    {
      "key": "e_1_3_1_29_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja00052a043"
    },
    {
      "key": "e_1_3_1_30_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja00173a031"
    },
    {
      "key": "e_1_3_1_31_2",
      "unstructured": "This assumes that the Fe-O-O-Fe torsion angle is zero. \u03b3 would be even smaller for nonzero torsion angles."
    },
    {
      "key": "e_1_3_1_32_2",
      "unstructured": "T. C. Brunold N. Tamura N. Kitajima Y. Moro-oka"
    },
    {
      "key": "e_1_3_1_32_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/ja980129x"
    },
    {
      "key": "e_1_3_1_33_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi00211a039"
    },
    {
      "key": "e_1_3_1_33_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi973128a"
    },
    {
      "key": "e_1_3_1_34_2",
      "unstructured": "Single-letter abbreviations for the amino acid residues are as follows: A Ala; D Asp; E Glu; F Phe; H His; Q Gln; S Ser; and W Trp."
    },
    {
      "key": "e_1_3_1_35_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1007/s007750050310"
    },
    {
      "key": "e_1_3_1_35_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1006/jmbi.1997.0970"
    },
    {
      "key": "e_1_3_1_36_2",
      "doi-asserted-by": "publisher",
      "DOI": "10.1016/S0021-9258(18)47653-3"
    },
    {
      "key": "e_1_3_1_36_3",
      "doi-asserted-by": "publisher",
      "DOI": "10.1021/bi00046a028"
    },
    {
      "key": "e_1_3_1_36_4",
      "doi-asserted-by": "publisher",
      "DOI": "10.1007/s007750050180"
    },
    {
      "key": "e_1_3_1_36_5",
      "unstructured": ". Kinetics of diferric peroxo formation and decay in a QXXA ferritin (human H) differ from either the QXXD (frog M) or QXXS (frog H); no information is available for the QXXQ ferritin (maize). In the Escherichia coli ferritins EXXH occurs instead of QXXD QXXA or QXXS as the ligands to the second Fe and the Fe is retained at the diiron site ["
    },
    {
      "key": "e_1_3_1_36_6",
      "doi-asserted-by": "publisher",
      "DOI": "10.1016/S0014-5793(98)00867-9"
    },
    {
      "key": "e_1_3_1_37_2",
      "unstructured": "We thank K. Herlihy G. W. Small and A. R. Tipton for assistance in the preparation of the ferritin samples and B. G. Fox for communicating results before publication. Supported in part by grants from NIH (GM-45205 to J.E.P.-H. GM-47295 and GM-58778 to B.H.H. and D.E.E. DK-20251 to E.C.T.). X-ray absorption data were measured at Stanford Synchrotron Radiation Laboratory which is operated by the U.S. Department of Energy (DOE) Office of Basic Energy Sciences with additional support from NIH National Center for Research Resources and the DOE Office of Biological and Environmental Research."
    }
  ],
  "container-title": "Science",
  "original-title": [],
  "language": "en",
  "link": [
    {
      "URL": "https://www.science.org/doi/pdf/10.1126/science.287.5450.122",
      "content-type": "unspecified",
      "content-version": "vor",
      "intended-application": "similarity-checking"
    }
  ],
  "deposited": {
    "date-parts": [
      [
        2024,
        1,
        13
      ]
    ],
    "date-time": "2024-01-13T09:53:46Z",
    "timestamp": 1705139626000
  },
  "score": 1,
  "resource": {
    "primary": {
      "URL": "https://www.science.org/doi/10.1126/science.287.5450.122"
    }
  },
  "subtitle": [],
  "short-title": [],
  "issued": {
    "date-parts": [
      [
        2000,
        1,
        7
      ]
    ]
  },
  "references-count": 54,
  "journal-issue": {
    "issue": "5450",
    "published-print": {
      "date-parts": [
        [
          2000,
          1,
          7
        ]
      ]
    }
  },
  "alternative-id": [
    "10.1126/science.287.5450.122"
  ],
  "URL": "http://dx.doi.org/10.1126/science.287.5450.122",
  "relation": {},
  "ISSN": [
    "0036-8075",
    "1095-9203"
  ],
  "subject": [],
  "container-title-short": "Science",
  "published": {
    "date-parts": [
      [
        2000,
        1,
        7
      ]
    ]
  }
}