ARTICLE Quantifying the strength of heterointeractions among receptor tyrosine kinases from different subfamilies: Implications for cell signaling Received for publication, March 30, 2020, and in revised form, May 20, 2020 Published, Papers in Press, May 27, 2020, DOI 10.1074/jbc.RA120.013639 Michael D. Paul1,2, Hana N. Grubb1 , and Kalina Hristova1,2,3,* From the 1Institute for NanoBioTechnology, 2Program in Molecular Biophysics, and 3Department of Materials Science and Engineering, Johns Hopkins University, Baltimore, Maryland, USA Edited by Alex Toker Receptor tyrosine kinases (RTKs) are single-pass membrane been shown to play important and diverse roles, such as amplifying proteins that control vital cell processes such as cell growth, sur- or inhibiting downstream signaling (28–34), increasing the signal vival, and differentiation. There is a growing body of evidence that diversity (35–38), or providing signaling back-up (27). RTKs from different subfamilies can interact and that these It is also becoming increasingly apparent that unliganded diverse interactions can have important biological consequences. RTK dimers can form in physiological contexts and can have However, these heterointeractions are often ignored, and their important effects, especially in disease (39–41). RTK overex- strengths are unknown. In this work, we studied the heterointer- pression has been linked to many cancers, as it promotes RTK actions of nine RTK pairs, epidermal growth factor receptor interactions even in the absence of ligand (17, 18). Interestingly, (EGFR)–EPH receptor A2 (EPHA2), EGFR–vascular endothelial some of these cancers are associated with not just increased growth factor receptor 2 (VEGFR2), EPHA2–VEGFR2, EPHA2– fibroblast growth factor receptor 1 (FGFR1), EPHA2–FGFR2, RTK expression, but loss of ligand as well. It can be expected EPHA2–FGFR3, VEGFR2–FGFR1, VEGFR2–FGFR2, and that such conditions (high RTK expression and no ligand) also VEGFR2–FGFR3, using a FRET-based method. Surprisingly, we favor the formation of heterodimers of unrelated RTKs, as found that RTK heterodimerization and homodimerization these RTKs typically do not share ligands. However, it is cur- strengths can be similar, underscoring the significance of RTK rently not clear how likely cross-subfamily heterointeractions heterointeractions in signaling. We discuss how these heteroin- are to occur, and it is thus difficult to say how significant a role teractions can contribute to the complexity of RTK signal trans- they play in RTK signaling. Indeed, heterointeraction strengths duction, and we highlight the utility of quantitative FRET for for RTKs from different subfamilies are unknown, as prior probing multiple interactions in the plasma membrane. work has had limited quantification capabilities. In this work, we study the heterointeractions of nine RTK pairs (EGFR–EPHA2, EGFR–VEGFR2, EPHA2–VEGFR2, The 58 human receptor tyrosine kinases (RTKs) are grouped EPHA2–FGFR1, EPHA2–FGFR2, EPHA2–FGFR3, VEGFR2– into 20 subfamilies based on the homology of their N-terminal FGFR1, VEGFR2–FGFR2, and VEGFR2–FGFR3) using a fluo- extracellular (EC) domains (1–4). They are single-pass membrane rescence-based method: fully quantified spectral imaging Förster receptors, which are activated upon dimerization and phosphoryla- resonance energy transfer (FSI-FRET). The FSI-FRET method tion of tyrosines in their C-terminal intracellular (IC) kinase reports on the occurrence of specific interactions and yields the domains (3, 5–8). Adaptor proteins bind to the phosphorylated strength of the interactions (1). We advance this FRET methodol- tyrosines, leading to the initiation of downstream signaling cascades ogy to also enable the determination of the oligomer size (e.g. that control cellular processes such as growth, motility, survival, dimer, trimer, tetramer) of the heterointeraction. For three of the – and differentiation (9 14). Moreover, RTKs are often dysregulated RTK pairs, we demonstrate that they form dimers in the absence – in many growth disorders and diseases, such as cancer (15 20). of ligand, and we quantify their heterodimerization strengths. Whereas RTKs are best known for forming signaling homo- Surprisingly, we find that the heterodimerization and homodime- dimers, they are also known to interact with many other binding rization strengths are similar. This suggests that the heterodimers partners, including other RTKs from the same subfamily, G pro- are likely to form under physiological and pathological condi- – – tein coupled receptors, and cell adhesion molecules (21 26). Fur- tions. Furthermore, we show that the presence of ligands modu- thermore, it is becoming clear that they are capable of interacting lates the abundance of the heterodimers. This work underscores with RTKs from unrelated subfamilies. A recent review of the liter- the significance of heterointeractions in RTK signaling and high- ature identified nearly 100 studies that support the idea that RTKs lights a method that can be used to study a multitude of interac- from different subfamilies can interact (27). Despite these studies, tions in the plasma membrane of live cells. heterodimerization across subfamilies has largely been ignored, both in conceptual models of RTK signaling and in the interpreta- Results tion of RTK signaling data. However, these heterodimers have Heterointeraction models In most cases, RTKs are thought to signal as homodimers * For correspondence: Kalina Hristova, [email protected]. (2, 3). Accordingly, if two different RTKs, X and Y, were to 7KLVLVDQ2SHQ$FFHVVDUWLFOHXQGHUWKH &&%<OLFHQVH J. Biol. Chem. (2020) 295(29) 9917–9933 9917 © 2020 Paul et al. Published under exclusive license by The American Society for Biochemistry and Molecular Biology, Inc. RTK heterointeractions Figure 1. RTK models. Shown are cartoon representations of the different RTK interaction models (A–C) and RTK constructs used in the study (D) (not drawn to scale and not meant to represent a mechanism of dimerization (e.g. the TM domains might not be directly interacting)). A, heterodimer model. B, heterotrimer model. C, heterotetramer model. D, fluorescent proteins (a FRET pair) are attached to the C termini of the RTK constructs used in this work. interact, the obvious assumption is that they would form a het- association constants can be written in terms of the concentra- erodimer, XY. However, there is no experimental evidence that tions of the monomers and dimers. excludes the possibility of higher-order heterooligomers (e.g. ½ trimers, tetramers) forming. For example, as each RTK explores XX KX ¼ (Eq. 1) a monomer-dimer equilibrium, it is possible that a dimer, YY, ½X 2 and a monomer, X, interact to form a trimer, XYY. It is also ½ possible that two dimers interact to form a tetramer, XXYY. YY KY ¼ (Eq. 2) These cases are illustrated in Fig. 1. Here, we develop physical- ½Y 2 chemical models for these cases, which can be used to identify ½XY the model that best describes experimental data. K ¼ (Eq. 3) XY ½X ½Y Assuming that the total concentration of each RTK is Heterodimers (Fig. 1A) constant—an assumption that is valid for the conditions The heterodimer case has been considered previously by under which the FRET experiments are performed (43)— Del Piccolo et al.(42). Even in this simplest case, three then simple equations for mass conservation can be written coupled reactions are needed to describe the fact that both as follows. X and Y form homodimers, XX and YY, in addition to the ½¼ ½1 ½1½ heterodimer, XY. Xtotal X 2XX XY (Eq. 4) K ½¼ ½1 ½1½ ½X 1½X $X ½XX Ytotal Y 2YY XY (Eq. 5) K – ½Y 1½Y $Y ½YY By substituting in the values of KX and KY from Equations 1 2, the total concentrations can be written in terms of the mono- ½ ½KXY ½ X 1 Y $ XY mer and heterodimer concentrations as follows. À Reactions 1 3 ½ ½2 ½ ½¼Xtotal X 12KX X 1 XY (Eq. 6) Here, KX and KY are the homodimer association constants, ½¼½1 K ½21½ and KXY is the heterodimer association constant. These three Ytotal Y 2 Y Y XY (Eq. 7) 9918 J. Biol. Chem. (2020) 295(29) 9917–9933 RTK heterointeractions These are quadratic equations which can be solved for in The equations for mass conservation can be written as terms of the monomer concentration. follows. pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ½¼½1 ½1½ À11 1 À 8K ð½XY À ½ÞX Xtotal X 2XX XYY (Eq. 14) ½X ¼ X total (Eq. 8) 4KX ½ ½½ ½¼Ytotal Y 12YY12 XYY (Eq. 15) qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiÀÁ À 1 À ½À ½ 1 1 8KY XY Ytotal and these can be rewritten in terms of the association ½Y ¼ (Eq. 9) 4KY constants. ½ ½2 ½ Last, the concentration of the heterodimer can be written in ½¼Xtotal X 12KX X 1 XYY (Eq. 16) terms of the monomer concentrations and the heterodimer ½ 2 ½ association constant by rearranging Equation 3 as follows. ½¼Ytotal Y 12KY½Y 12 XYY (Eq. 17) ½½ ½XY¼KXY X Y (Eq. 10) These are quadratic equations which can be solved for in terms of [X] and [Y]. By substituting the values of [X] and [Y] from Equations 8 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi and 9 into Equation 10, we arrive at an equation where K À 1 À ð½À ½Þ XY ½¼ 1 1 8KX XYY Xtotal and[XY]aredefinedintermsof[X ], [Y ], K ,andK . X (Eq. 18) total total X Y 4KX As described previously (1, 6, 42, 44), quantitative FRET qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiÀÁ experiments can directly measure [Xtotal]and[Ytotal], and ½ À11 1 À 8KY 2 XYY À ½Ytotal the FRET efficiency. FRET efficiency is a function of [XY], as ½Y ¼ (Eq. 19) shown in Equation 32 (see “Materials and methods”).