What an SPR Instrument Actually Measures

The one-sentence version#

A surface plasmon resonance instrument measures the refractive index of a thin slab of liquid sitting within roughly 200 nanometres of a gold surface, and it does so a few times per second. Everything else — kinetics, affinity, concentration, epitope maps — is inference layered on top of that single measurement.

That sentence is worth memorising, because most SPR mistakes are made by people who believe the instrument measures binding. It does not. It measures refractive index. Binding is one thing that changes refractive index near a surface. A temperature fluctuation is another. A different buffer arriving in the flow cell is another. Protein sticking non-specifically to the dextran is another. The instrument cannot tell these apart; only your experimental design can.

Why label-free matters#

Before SPR, measuring an interaction usually meant attaching something to one of the partners: a radioisotope, a fluorophore, an enzyme. Labels work, but they carry three costs. They take time and material to install. They can perturb the very binding you are trying to measure — a fluorophore on a lysine near the interface is not a neutral observer. And they generally give you an endpoint, not a time course, so you learn the affinity but not how fast the complex forms or falls apart.

The first demonstration that SPR could watch an immunological reaction without any label at all came from Liedberg, Nylander and Lundström in 1983, who adsorbed human IgG onto a 60 nm silver film and detected anti-IgG arriving from solution (Liedberg et al., 1983). The same group had used the effect for gas sensing the year before (Nylander et al., 1982). What made the 1983 experiment consequential was not sensitivity — it was that the signal appeared in real time, as a curve rather than a point.

Anatomy of an experiment#

Almost every SPR experiment, from a first-year practical to a fragment screen against a kinase, has the same five acts. Learn the shape once and you can read any sensorgram in the literature.

1. Immobilisation. One partner — by convention the ligand — is attached to the sensor surface (Johnsson et al., 1991). This is a chemistry step and it happens once per surface.
2. Baseline. Running buffer flows over the surface until the signal is flat. A drifting baseline here means the experiment is not ready.
3. Association. A solution containing the analyte is injected. If it binds, mass accumulates near the surface and the response climbs.
4. Dissociation. The injection ends and buffer flows again. Complex that falls apart is washed away, and the response decays.
5. Regeneration. A short pulse of something harsh — usually low pH — strips remaining analyte without destroying the ligand, returning the surface to baseline so the next concentration can be run (Karlsson et al., 1994).

The two ends of that curve carry different information, and this is the single most useful thing to internalise early. The association phase contains information about both rate constants. During injection, complex is forming and falling apart simultaneously, so the observed approach to steady state has an exponential constant k_obs = k_a·C + k_d (O’Shannessy et al., 1993). The dissociation phase contains information about k_d only, because with analyte absent from the bulk there is nothing left to associate. This is why a clean, long dissociation phase is worth more than another two analyte concentrations (Karlsson & Fält, 1997).

What one response unit means#

Instrument manufacturers report response in response units. On Biacore-lineage instruments 1 RU corresponds to a shift of the resonance angle of 10^-4 degrees (Jönsson et al., 1991). That is a definition, not a measurement — it tells you nothing biological on its own. The biologically useful statement is the calibration that connects it to surface mass.

Stenberg and colleagues did that calibration the honest way, by flowing radiolabelled protein over a surface and counting how much stuck while simultaneously watching the SPR shift (Stenberg et al., 1991). The result is the relation every SPR user quotes:

The underlying physics is the de Feijter relation: the refractive index of a thin adsorbed layer rises linearly with the mass concentration in it, with slope dn/dc (Feijter et al., 1978). Jung and colleagues worked out how a thin film of known thickness and refractive index maps onto an SPR shift, including the exponential weighting that means material further from the gold counts for less (Jung et al., 1998). Two molecules of the same mass but different dn/dc produce different signals. That is why comparing SPR responses between a protein and a DNA oligo of the same molecular weight is not meaningful without correction.

What questions SPR can answer#

SPR has since been applied far beyond this list — to gases, cells, vesicles, polymers and surfaces of every kind (Homola, 2008) — but the logic of the table does not change. Note that only the first row is easy. Everything below it requires design decisions that the instrument will not make for you and will not warn you about. The remainder of this site is, in effect, a long answer to the question "what could go wrong in rows two through six?"

Sources cited on this page

Listed alphabetically. Each badge records whether the bibliographic record was confirmed against Crossref. unverified marks a real, deliberately chosen source whose volume and page numbers we have not yet machine-checked — it is not a comment on the science.

• Day et al., 2002Y. S. N. Day, C. L. Baird, R. L. Rich, D. G. Myszka (2002). Direct comparison of binding equilibrium, thermodynamic, and rate constants determined by surface- and solution-based biophysical methods. Protein Science 11, 1017–1025. doi:10.1110/ps.4330102 unverified
  The head-to-head comparison of SPR against solution methods on the same systems. The primary source for the claim that a well-run surface measurement agrees with solution thermodynamics.
• Feijter et al., 1978J. A. de Feijter, J. Benjamins, F. A. Veer (1978). Ellipsometry as a tool to study the adsorption behavior of synthetic and biopolymers at the air–water interface. Biopolymers 17, 1759–1772. unverified
  Source of the de Feijter equation relating adsorbed mass to refractive index change via dn/dc — the reason a response unit can be read as a mass.
• Homola, 2008J. Homola (2008). Surface plasmon resonance sensors for detection of chemical and biological species. Chemical Reviews 108, 462–493. doi:10.1021/cr068107d unverifiedreview
  Cited only for statements about the breadth of the field, never for a specific experimental result.
• Johnsson et al., 1991B. Johnsson, S. Löfås, G. Lindquist (1991). Immobilization of proteins to a carboxymethyldextran-modified gold surface for biospecific interaction analysis in surface plasmon resonance sensors. Analytical Biochemistry 198, 268–277. doi:10.1016/0003-2697(91)90424-R corrected
  Note on this citation. This paper is widely cited with the page range 168–277, which is a transposition error that has propagated through several secondary sources. The Crossref record for this DOI gives 268–277, which is also what volume 198 of the journal actually contains. We use the corrected range.
  The EDC/NHS amine-coupling protocol, still the default immobilisation chemistry three decades later.
• Johnsson et al., 1995B. Johnsson, S. Löfås, G. Lindquist, Å. Edström, R.-M. Müller Hillgren, A. Hansson (1995). Comparison of methods for immobilization to carboxymethyl dextran sensor surfaces by analysis of the specific activity of monoclonal antibodies. Journal of Molecular Recognition 8, 125–131. doi:10.1002/jmr.300080122 unverified
  Measures surface activity across coupling chemistries on the same antibody — the primary source for the claim that amine coupling inactivates a substantial fraction of ligand.
• Jönsson et al., 1991U. Jönsson, L. Fägerstam, B. Ivarsson, B. Johnsson et al. (1991). Real-time biospecific interaction analysis using surface plasmon resonance and a sensor chip technology. BioTechniques 11, 620–627. unverified
  The paper describing the first commercial SPR biosensor as an integrated system — optics, microfluidics and sensor chip together. The primary source for what that instrument was and did.
• Jung et al., 1998L. S. Jung, C. T. Campbell, T. M. Chinowsky, M. N. Mar, S. S. Yee (1998). Quantitative interpretation of the response of surface plasmon resonance sensors to adsorbed films. Langmuir 14, 5636–5648. doi:10.1021/la971228b unverified
  The reference treatment of how a thin adsorbed film of known thickness and refractive index maps onto an SPR shift, including the exponential weighting of the evanescent field.
• Karlsson et al., 1991R. Karlsson, A. Michaelsson, L. Mattsson (1991). Kinetic analysis of monoclonal antibody–antigen interactions with a new biosensor based analytical system. Journal of Immunological Methods 145, 229–240. doi:10.1016/0022-1759(91)90331-9 verified
  The paper that established that a commercial SPR instrument could return separate association and dissociation rate constants, not merely an affinity.
• Karlsson et al., 1994R. Karlsson, H. Roos, L. Fägerstam, B. Persson (1994). Kinetic and concentration analysis using BIA technology. Methods 6, 99–110. doi:10.1006/meth.1994.1013 verified
• Karlsson & Fält, 1997R. Karlsson, A. Fält (1997). Experimental design for kinetic analysis of protein–protein interactions with surface plasmon resonance biosensors. Journal of Immunological Methods 200, 121–133. doi:10.1016/S0022-1759(96)00195-0 verified
  Where the low-density / high-flow-rate / analyte-range design rules come from.
• Liedberg et al., 1983B. Liedberg, C. Nylander, I. Lundström (1983). Surface plasmon resonance for gas detection and biosensing. Sensors and Actuators 4, 299–304. doi:10.1016/0250-6874(83)85036-7 verified
  The founding paper of SPR biosensing: IgG adsorbed on a silver film, anti-IgG detected from solution, no label.
• Myszka, 1999D. G. Myszka (1999). Improving biosensor analysis. Journal of Molecular Recognition 12, 279–284. unverified
  Origin of double referencing and blank-injection subtraction as standard practice.
• Nylander et al., 1982C. Nylander, B. Liedberg, T. Lind (1982). Gas detection by means of surface plasmon resonance. Sensors and Actuators 3, 79–88. doi:10.1016/0250-6874(82)80008-5 verified
  SPR used as a transducer for the first time — for gas, one year before biomolecules.
• O’Shannessy et al., 1993D. J. O’Shannessy, M. Brigham-Burke, K. K. Soneson, P. Hensley, I. Brooks (1993). Determination of rate and equilibrium binding constants for macromolecular interactions using surface plasmon resonance: use of nonlinear least squares analysis methods. Analytical Biochemistry 212, 457–468. doi:10.1006/abio.1993.1355 verified
  The case for fitting the sensorgram directly rather than linearising it.
• Stenberg et al., 1991E. Stenberg, B. Persson, H. Roos, C. Urbaniczky (1991). Quantitative determination of surface concentration of protein with surface plasmon resonance using radiolabeled proteins. Journal of Colloid and Interface Science 143, 513–526. doi:10.1016/0021-9797(91)90284-F verified
  The calibration behind the response unit. Radiolabelled protein was used to tie SPR angle shift to actual surface mass, giving the ~1 RU ≈ 1 pg mm⁻² relation.
• Zhao et al., 2011H. Zhao, P. H. Brown, P. Schuck (2011). On the distribution of protein refractive index increments. Biophysical Journal 100, 2309–2317. unverified
  Measured spread of dn/dc across proteins; the basis for treating ~0.185 mL g⁻¹ as near-universal and for knowing how wrong that assumption can be.