How Do Instructors Explain The Mechanism by which ATP Drives Unfavorable Processes?

INTRODUCTION AND BACKGROUND

A History of Educational Research Involving ATP

It could be argued that no scientific topic has been more widely used and less well understood than energy. Although it is difficult to define (Feynman et al., 2011), energy plays a critical role within individual disciplines and as a crosscutting concept that transcends disciplines (National Research Council, 2012; Cooper and Klymkowsky, 2013a). Given its central position, one might imagine that the concept of energy could provide a unifying framework to support connections between scientific disciplines. However, most disciplines have adopted siloed approaches to thinking about energy, contributing to the unfortunate fact that energy is a notoriously difficult topic for both students and instructors to grasp in a useful and productive way (Goldring and Osborne, 1994; Barak et al., 1997; Barker and Millar, 2000; Cooper and Klymkowsky, 2013a). These difficulties are exacerbated when studying molecular-level phenomena, which involve unfamiliar symbols and is often based on nonintuitive ideas, for example, “hydrophobic interactions” are attractive and not repulsive (Cooper and Klymkowsky, 2013a). One of the most well-documented, problematic ideas about energy is the belief that breaking chemical bonds or noncovalent interactions releases “energy” into the surroundings when, in fact, the opposite is true. For nearly 50 years, researchers have documented the widespread presence of this misconception and tried to develop instructional approaches that support a more accurate understanding of energy (Novick, 1976; Johnstone and Mahmoud, 1980; Hapkiewicz, 1991; Boo, 1998; Barker and Millar, 2000; Teichert and Stacy, 2002; Galley, 2004; Dreyfus et al., 2014; Green et al., 2021; VandenPlas et al., 2021). According to this work, one molecule appears to contribute significantly to the issue: adenosine triphosphate or ATP, a key molecule in biological systems. ATP plays a central role in cellular metabolism. It is frequently described as the “energy currency” of the cell, storing energy so it may later be used to allow thermodynamically unfavorable processes to proceed (Storey, 1992; Freeman et al., 2019; Urry et al., 2020). In this paper, we refer to ATP as an energy “carrier” or “source” interchangeably. We also frequently discuss driving unfavorable processes as a more general term that includes driving unfavorable reactions (a more specific process that can be discussed at the molecular/chemical level).

The role of ATP as the link between energy storage and use in the cell was first proposed by Lipmann (1941), which contributed to his Nobel Prize in 1953. Lipmann discussed the “energy-rich” bond between phosphate groups and noted that the hydrolysis of these groups could release a large amount of heat. Unintentionally, this led to the logical, but incorrect, simplification that it is the breaking of this energy-rich bond in ATP which releases energy. This idea directly contradicts the fact that energy is required to overcome (to break) the stabilizing attractive interactions of covalent bonds and noncovalent interactions.

Thus, in chemistry and biology courses, we may be offering contradictory descriptions of the same phenomena – a situation that can result in didaskalogenic (i.e., instruction-induced) misunderstandings that prevent students from developing a coherent and productive understanding of energy. Consider this quote regarding bonding and energy from Kohn et al. (2018), in which an undergraduate student concurrently enrolled in chemistry and biology courses said: “I know for biology what [the instructor] wants us to say and then for chemistry what we have to say”. With all the confusion surrounding energy, bonds, and ATP, it is worth considering whether there is a more useful way to talk about ATP. More specifically, how can we talk about ATP in a way that is both productive in biology and in alignment with student understanding of chemical principles?

In the 1970s, Novick (1976) published one of the first studies documenting the presence of the “misconception” about bonds and energy among university students. He found that when asked to provide a molecular interpretation of how “fats supply energy to the body”, the majority of students suggested that energy was stored within chemical bonds that could be released when the bond was broken. Specifically, these students often referenced the bonds within ATP. For example, one student explained that “the breakdown of fats produces a source of energy in the form of molecules (ATP), having special bonds which, when broken, release energy.” Since then, other studies have expanded upon this work, providing additional evidence for confusion among students regarding the energy associated with the functional roles of ATP in biological systems (Johnstone and Mahmoud, 1980; Hapkiewicz, 1991; Teichert and Stacy, 2002; Galley, 2004).

More recently, researchers have moved beyond documenting student misunderstandings about bond energies and ATP, to proposing instructional solutions to these issues (Dreyfus et al., 2014; Green et al., 2021; VandenPlas et al., 2021). The common thread across these efforts has been to explain how and why energy is released during ATP hydrolysis (i.e., the reaction of ATP with water to form ADP and inorganic phosphate), by leveraging bond energies more appropriately. Specifically, the focus of these efforts lies on how instructors can provide consistency in the way bond energies are discussed across disciplines or, at the very least, help students avoid developing misconceptions about how and why this reaction releases energy. The results show that, with a clear focus on these ideas, students can develop a more canonical understanding of why ATP hydrolysis is accompanied by release of energy (i.e., more energy is released upon bond formation, due to electrostatic-attractive forces, than is required for bond breaking). However, focusing on the energy release of ATP hydrolysis may generate its own unproductive ideas, because in cellular systems, a negligible amount of energy is actually “released” to the surroundings. Rather, these systems have evolved such that energy is transferred (or used) both within and among systems and their components. In fact, many ATP-dependent processes involve phosphorylation of a molecule or protein and, thus, do not involve an ATP hydrolysis step, for example, glutamine synthesis (Liaw and Eisenberg, 1994; Moreira et al., 2017) or Ca²⁺ protein pumps (Inesi, 1985; Toyoshima and Inesi, 2004; Møller et al., 2005; Das et al., 2017). For those reactions in which ATP hydrolysis is involved, for example the F0-ATPase proton pump, the hydrolysis step does not provide the major driving force for the process (Feldman and Sigman, 1982; Johnson, 1985; Nakamoto et al., 2008; Kiani and Fischer, 2013; Geeves, 2016; Prieß et al., 2018).

Perhaps even more importantly, if we invoke ATP hydrolysis by emphasizing energy release, this approach provides no mechanism for how the energy is transferred among components in biological systems. If ATP hydrolysis occurred in isolation, it would lead to an increase in the overall temperature of the system – an inefficient and nonspecific way to speed up favorable reactions and to drive unfavorable processes (to say nothing of the effects on the organism). We are not the first to note this disconnect. As early as 1970, researchers outside of the educational field noted that “the original concept put forward by Lipmann was ill-founded and that its effect is to divert attention from the genuine problem of the mechanism of events in which ATP takes part” (Banks and Vernon, 1970). Carusi (1992) noted that most textbooks provide no explanation for how the energy released from “high-energy” bonds is coupled to the unfavorable reactions they drive, lamenting that without this mechanism “the student is left in the dark waiting to learn how the free energy available from these compounds actually is transferred and utilized”. While this paper is over 30 y old, and the mechanisms by which ATP drives these reactions has perhaps made it into some textbooks, current research shows the way ATP is discussed and represented (most often as the hydrolysis of ATP) in commonly-used textbooks is ripe with issues (Yang, 2023), emphasizing the need for additional work investigating instructional practices and materials associated with ATP.

Our work aims to address this educational gap by exploring common themes in the mechanisms by which ATP drives unfavorable processes. Thus, we leverage mechanistic reasoning, which can be used as not only an explanatory and predictive thinking strategy, but also an instructional emphasis intended to provide students with a productive understanding of the roles of ATP in biological systems.

Mechanistic Reasoning

There is emerging consensus in science education literature that mechanistic reasoning – that is, how and why phenomena occur – should be practiced by students and implemented in course design and instruction (National Research Council, 2012; Cooper, 2015; Glennan, 2017; Schwarz et al., 2017). Not only does mechanistic reasoning provide students with ways to explain and predict phenomena, but also, as recently shown, assessments incorporating mechanistic reasoning are more equitable than more traditional-test items that focus on procedural/mathematical skills or rote knowledge (Ralph et al., 2022). Thus, we argue that explicitly discussing the mechanisms by which biological systems act would be useful for all students, and specifically introductory biology students (van Mil et al., 2013; Klymkowsky, 2021). Many molecular biologists and biochemists spend their time searching for and/or predicting cellular mechanisms, thus providing students the opportunity to engage in this practice in their courses is a useful endeavor. For these reasons, we posit that research intended to help students develop a mechanistic understanding of how ATP functions in cellular systems, is a valuable area of investigation.

In earlier publications (Noyes et al., 2022; Franovic et al., 2023), we used the mechanistic reasoning framework proposed by Russ et al. (2008) and elaborated on by Krist et al. (2019). According to Krist et al., mechanistic reasoning is an epistemic heuristic, or thinking strategy, that can be leveraged across all science disciplines and, when used appropriately, results in powerful (testable) predictions. Russ, Krist, and others (Machamer et al., 2000; Braaten and Windschitl, 2011; van Mil et al., 2013) broadly define mechanistic reasoning as reasoning about how and why phenomena occur by identifying relevant (often lower scalar level) entities, the properties/activities of those entities, and how behaviors or interactions of the entities link together and give rise to the phenomenon under consideration. However, while this approach is successful for relatively simple phenomena (Becker et al., 2016; Crandell et al., 2020; Franovic et al., 2023), such as phase changes or chemical reactions, no published work to our knowledge has investigated its effectiveness for more complex biological phenomena. These complex phenomena often span several scalar levels and require ideas from more than one discipline, making it more difficult to identify the level of depth at which an explanation is appropriate (Haskel-Ittah, 2023). Further, recent work in our group has shown that students struggle to reason mechanistically about complex phenomena, such as that of protein-ligand binding at both the biochemical and population levels (S.M., F.C.G.-C., de L.J., N.K., B.D., B.-F.E., K.J., M.R.L., P.-G.E., C.M.M., L.T.M., S.C.V., & S.J.R., unpublished data).

One of the challenges learners encounter when crafting an explanation is determining the necessary and sufficient depth for their explanation. For example, if we consider the role of ATP in driving unfavorable processes, there are a number of valid approaches that can be used: 1) we might think macroscopically about the overall reaction and how the hydrolysis of ATP releases a larger amount of energy than that required to drive the unfavorable reaction – an approach that typically involves thermodynamic calculations using Gibbs free energy; 2) we might approach the problem at the molecular level, invoking the idea of ATP as a phosphorylating agent, leading to a more reactive common intermediate, which then reacts to produce the product; 3) we might use a systemic approach to explain how the cell generates and maintains a relatively high concentration of ATP, thereby increasing the rate of collisions between ATP and the reactant, pushing a sequence of reactions forward; 4) we might consider the subatomic level, that is, the electron distributions of the participating entities and how a nucleophilic substitution reaction with ATP and a substrate occurs as a result of electrostatic forces and interactions (an approach that may be more suitable in organic chemistry and more advance biochemistry courses); or 5) we might consider reactions in which the three-dimensional structure of an enzyme is altered by the binding of ATP as opposed to the situation when adenosine diphosphate is bound. Deciding the appropriate level and type of explanatory depth can depend on several factors including, for example, the content knowledge available to the explainer, that of the student, the discipline and course level in which the question is posed, or simply the question being asked.

In undergraduate science courses, the weight of identifying what to explain mechanistically (i.e., deciding sufficient depth) and in turn, what to expect of students typically falls on the instructors. How then do these instructors think about the mechanisms by which ATP drives unfavorable processes? What mechanisms do they emphasize during instruction when explaining how ATP “works”? These were the questions we were interested in while exploring in this study.

PURPOSE AND RESEARCH QUESTIONS

We interviewed instructors who had taught undergraduate courses in biology, chemistry, and/or biochemistry with the aim of gaining insight into instructor and disciplinary-expert understandings and explanations about energy, in particular how ATP provides the energy needed to drive unfavorable processes (i.e., how does ATP “work”). On its own, the statement “ATP drives unfavorable processes” is an explanatory black box, or a unit within an explanation that remains unexplained (Haskel-Ittah, 2023), which might be unpacked using different approaches (such as explaining the mechanism of energy release or energy transfer) during instruction. Black boxes have historically carried a negative connotation; however, in this work, we recognize the necessity and utility of explanatory black boxes, particularly for complex phenomena which consist of many “mechanisms within mechanisms”, to explain all of which would be tedious and unproductive (Haskel-Ittah, 2023). The findings we discuss here can support the development of alternative curricular materials (such as formative assessments) that better support students’ mechanistic understanding of the role of ATP in driving unfavorable processes, as well as their understanding of energy.

While our aim was to investigate how instructors explain and think about the mechanism(s) by which ATP drives unfavorable processes, most participants also revealed personal experiences, providing affective insights about teaching ATP in their course(s). Thus, we address three different research questions, the first two capturing our initial goal and the third as an artifact (bonus) that emerged from these rich discussions:

1. What ideas did instructors leverage when discussing ATP?

2. In what ways do instructors use these ideas to discuss how ATP drives unfavorable processes?

3. What teaching experiences did molecular biology (MB) instructors share regarding ATP in their course(s)?

In this analysis, we do not intend to portray any instructors as superior or inferior based on how they unpacked this phenomenon during their interviews. Rather, we share several excerpts which sparked deep discussions among us and potential avenues for future research in biology education.

METHODS

Participants

Between 2018 and 2020, we conducted semistructured interviews with 15 instructors of undergraduate courses in biology, chemistry, and/or biochemistry. We used a convenience sampling approach to select these instructors, soliciting participation from instructors with whom we were familiar. All participants consented to the interview in accordance with our institutional review board protocol. The instructors were not compensated for participating in these interviews. To protect the identity of each participant, we use pseudonyms and gender-neutral pronouns (they/them) throughout this manuscript. Table 1 provides the pseudonym, primary course taught, primary research (if applicable), interviewer, and the interview medium. All instructors were from one large, public, research-intensive midwestern university, except Dracaena, who was from a public, midsized (∼25,000 students), midwestern university.

Instructors listed according to discipline (green^b = biology, blue^bc = biochemistry, gray^c = chemistry)

Data Collection

Authors C.F. or K.N. conducted semistructured interviews, which allowed for the inclusion of our interests as well as authentic ideas from participants, either in-person or via the online meeting software Zoom (Table 1). In these interviews, we explored how the instructors think about sources of energy in their course/discipline, and the role(s) of ATP in such systems. To start each interview, we asked them about the courses they teach, and to describe how they think about energy in chemical and biological systems. We began with these general questions to ease into the more complex discussions that emerged when we asked how they think about the mechanism(s) by which ATP is used as an energy source (Figure 1 shows this general interview protocol, and Supplementary Figure S1 provides the entire protocol).

In this study, we focused on their explanations of how ATP “works.” Typically, this occurred in response to the question: how do you think about the mechanism by which ATP is used as an energy source? In addition to their thoughts about the mechanism, we asked whether/how they teach these ideas in their course(s). Some participants also discussed ATP and its role before our interview questions (unprompted).

We recorded the audio of each interview and transcribed them using either Express Scribe, software that facilitates human transcription of audio, or Otter, an automated transcription software. A second researcher read through each transcript while listening to the audio recordings to edit the transcript in order to confirm that it accurately captured the dialogue from each interview.

Positionality Statement

We intend to be as transparent as possible in presenting this work by stating our positions, which impacted the data collection, analysis, and interpretation. Authors C.F. and K.N. were graduate students when conducting all interviews, placing them as “subordinate” to the faculty participants; however, as interviewers, we recognize our position of power in knowing the questions and topics to be discussed. To mitigate the effects of this position, we spent the first part of each interview establishing trust with the participants and engaging in casual conversation before recording. Second, all authors believe in the importance of using a mechanistic-reasoning approach to teach and learn about scientific phenomena – it has been a driving force in various course redesign projects initiated by authors M.W.K. and M.M.C. (Cooper and Klymkowsky, 2013b; Klymkowsky et al., 2016; Cooper et al., 2019). Thus, we view mechanistic reasoning as a more productive way of thinking about science when compared with rote memorization or knowledge as a set of facts. In leading up to this study, we consulted the literature and thought deeply about the mechanisms of ATP in cellular systems. Thus, we had preexisting ideas about what the participants might discuss when we asked them to explain the mechanism by which ATP drives unfavorable processes and the degree to which they align with previously studied biological mechanisms. Specifically, the idea that ATP hydrolysis (i.e., the energy release mechanism) does not address the mechanism by which energy is transferred to the system to drive the unfavorable processes. These preexisting ideas consciously and subconsciously shaped our analysis and, therefore, should be acknowledged. We did, however, share this analysis with scholars outside of our group for input and/or other interpretations. Lastly, our disciplinary identities include a range of chemistry and biology backgrounds.

Data Analysis

Research Question 1.

To identify the different ideas instructors leveraged when discussing ATP, we analyzed the full interviews using MAXQDA 2020 (VERBI Software, 2019), a qualitative data analysis software. We began our analysis by first familiarizing ourselves with the data – that is, authors C.F., K.N., and M.M.C. read through each interview multiple times, recording notes and questions, and discussed our findings (Creswell, 2013). These discussions informed the open coding conducted by authors C.F. and N.W., in which they open-coded the interviews in sets of three, compared generated codes, and began coding scheme development. At this point in coding, we decided to narrow the scope of our analysis to focus solely on participant discussion of ATP; however, we continued to read through the entire interview of each participant to capture any unprompted discussions about ATP.

We continued refining the coding scheme using a constant comparative (Corbin and Strauss, 2008) approach until we felt that the codes represented the range of ideas expressed across all interviews (i.e., no additional codes were generated). The final coding scheme consisted of 19 codes (Supplementary Table S2). The length of coded segments varied based on what we deemed sufficient to capture the essence of how the instructor used that code. That is, we did not simply count the frequency of the term “bond energy”; rather, we read the transcripts in detail to identify sections in which the participants leveraged these ideas (and how). The majority of the codes are disciplinary content-related, as our protocol mainly focused on eliciting instructor explanations for how ATP functions (from both a chemical and biological perspective). However, some codes related to other aspects of the instructors’ experiences (rather than content). For example, the code teaching/learning captured instructor comments related to an experience regarding ATP in the classroom. We provide descriptions and examples of some of the most frequently occurring codes in Table 2 (descriptions and examples of all other codes are included in Supplementary Table S2). We discuss the importance of our codes in the results section of RQ1.

Using this scheme, two coders (C.F. and N.W.) independently coded each interview. Our initial agreement was 85.67%; however, the two authors met and discussed any disagreements until we reached 100% consensus. The codes reported in this manuscript are the consensus codes. The final codes were nonmutually exclusive and could appear any number of times in each interview.

Research Questions 2 and 3.

To address RQ2 and RQ3, we explored the participants’ discussions of how ATP drives unfavorable processes (RQ2) and the experiences that MB instructors shared when reflecting on their teaching of ATP (RQ3). The coding scheme we developed to answer RQ1 captured all ideas related to ATP that emerged throughout the entirety of each interview. We used a smaller subset of these codes to define the larger themes which more directly answer RQ2 and RQ3. For RQ2, we used codes that specifically indicated instructor explanations of 1) the mechanism of energy release (ATP hydrolysis and bond energy) and/or 2) the mechanism of energy transfer (phosphate transfer and common intermediate) as a guide in our selection of excerpts (of varying length) highlighting these two broader themes. To answer RQ3, we used the teaching/learning code to identify the experiences instructors shared regarding their teaching practices about how ATP provides energy in a coupled reaction. Eliciting these experiences was not an intentional aim of our protocol; however, we would be remiss to exclude these experiences when they emerged through such little prompting. It is likely that these experiences inform instruction (and, subsequently, student learning), making this an important and meaningful research question. Due to the dependence of RQ2 and RQ3 on the codes discussed in RQ1 results, we have included additional details about the data analysis of RQ2 and RQ3 in the corresponding results sections.

RESULTS

Research Question 1 – What Ideas did Instructors Leverage When Talking About ATP?

We coded all the ideas that participants mentioned when discussing ATP in each interview, resulting in 19 total codes (italicized in the body text; Supplemental Table 2 and Supplemental Figure S1). Our coding scheme captured the wide range of ideas mentioned when considering ATP; however, no instructor discussed every code, and most instructors only leveraged a smaller selection of these ideas. Additionally, given the nature of our coding scheme, some participants received the same code multiple times. Table 3 shows the presence and frequency of codes identified in each interview.

The codes are listed in order of occurrence, with codes that appeared in the most interviews at the top, and those that appeared in the fewest interviews at the bottom. The instructors are ordered based on their discipline (green^b = biology, blue^bc = biochemistry, gray^c = chemistry). Darker shading represents a higher frequency of occurrence.

While the majority of the codes relate to disciplinary concepts (e.g., bond energy, gradients), two of the most frequently occurring codes do not: teaching/learning and referencing disciplines. Rather, these codes address experiences regarding how ATP is taught or understood in the classroom (teaching/learning) and how their own or different disciplines might discuss ATP (referencing disciplines). While our interview protocol prompted instructors to consider their teaching practices, the prominence of these codes highlights the range of personal experiences instructors felt inclined to share about how ATP is discussed in their courses and the courses of other disciplines. This was particularly common among biologists reflecting on how their teaching aligns or differs with the concepts taught in chemistry. We discuss these experiences in more depth in RQ3.

Among the codes related to disciplinary concepts, participants discussed ideas like the production of ATP (ATP synthesis and gradients) and the relevance of equilibrium. However, some of the most popular codes correspond to two important mechanisms associated with ATP as an energy source (the primary interest of this study): 1) how energy is released and 2) how energy is transferred. Codes related to energy release, bond energy and ATP hydrolysis, were used extensively across disciplines, with all instructors discussing one or both of these ideas at some point in their interviews. On the other hand, the codes phosphate transfer (N = 7) and common intermediate (N = 6), related to the energy transfer mechanism, were present in many, but not all, of the interviews. Another frequently occurring code, coupled reactions, overlapped often with explanations for both mechanisms; this code captured more general statements about reaction coupling and did not necessarily contribute to explaining how or why either mechanism occurs.

Interestingly, there were some notable differences in the degree to which participants leveraged ideas related to energy transfer and energy release. For example, Eucalyptus^b discussed phosphate transfer during their interview but not ATP hydrolysis; Ficus^b talked about ATP hydrolysis but not common intermediate; Lily^b discussed ATP hydrolysis and common intermediate (each with high frequencies). We were interested in the instructors’ use of these codes (and therefore their adherence to mechanisms of energy transfer and/or energy release) in their explanations for how ATP drives unfavorable processes, which we address in RQ2.

Research Question 2 – In What Ways do Instructors Use These Ideas to Discuss How ATP Drives Unfavorable Processes?

To explore how instructors think about the mechanism by which ATP drives unfavorable processes, we used our coding from RQ1 to analyze the interview sections which reflected the ideas instructors leveraged when “unpacking” (or opening up) the black box of ATP “providing” energy. Based on the results of RQ1, we found that the two most popular mechanisms to unpack were: 1) energy release and 2) energy transfer. One instructor, Orchid^c, did not explain either mechanism in enough detail for us to confidently place them in one of the themes (even though they did receive one instance of the code bond energy); they were removed from the analysis for RQ2.

The instructors we interviewed discussed 1) how energy is released by explaining how ATP is hydrolyzed (ATP hydrolysis) and that the bonds formed are stronger than the bonds broken (bond energy), and/or 2) how ATP transfers energy via the transfer of a phosphoryl group to some other entity (based on the presence of the codes phosphate transfer and/or common intermediate). While some instructors only explained one mechanism, several explained both at some point in the interview, which we discuss by providing example excerpts below. In the excerpts, “[Editorializing]” indicates rephrasing (of the participant or interviewer).

Energy Release.

Excluding Orchid^c, all 14 instructors explained the mechanism of energy release at some point in their interviews. These explanations included ideas about ATP hydrolysis and/or bond energy, neither of which were included in our interview protocol (i.e., all the instructors brought up these ideas without us mentioning either of them explicitly), with a focus on how ATP hydrolysis releases energy and/or the idea that the bonds in ATP (and water) are less stable than those in ADP and inorganic phosphate. For example, Basil^c said, “when they often talk about it in biology, […] hey we break the […] ATP bond and energy comes out. And they completely ignore sort of the fact that there are a whole bunch of new products that get formed and that water is involved…” (codes: referencing disciplines and ATP hydrolysis), suggesting that water should be included when explaining the mechanism by which ATP drives unfavorable processes. Using the ideas of ATP hydrolysis and bond energy also encompass the ways in which recent publications have suggested teaching about ATP as a source of energy (Dreyfus et al., 2014; Green et al., 2021; VandenPlas et al., 2021). Figure 2 provides excerpts of three instructor explanations. In the interest of space, we condensed quotes by using ellipses in brackets ([…]), which indicate repetitive dialogue or filler words/sentences that did not contribute significantly to an understanding of the excerpt.

Energy Transfer.

While all participants highlighted the mechanism of energy release by discussing ATP hydrolysis and/or bond energy at some point, only nine participants explained a mechanism of energy transfer, which we identified as those who included the ideas (codes) phosphate transfer and/or common intermediate.Figure 3 provides excerpts and codes from three participants (one from each discipline) who explained energy transfer using one or both of these ideas. For example, Ivy^bc (Figure 3a) explained the mechanism by which ATP transfers energy, by leveraging the idea of ATP as “the universal phosphoryl group donor” (code: phosphate transfer) and its role in forming a phosphorylated intermediate (code: common intermediate). Unlike Ivy^bc, Ginkgo^c (physical chemistry) did not identify the role of ATP as a phosphorylating agent, but they emphasized the importance of energy being transferred mechanically or “like a bicycle gear”, forming a common, high-energy intermediate to drive unfavorable reactions (code: common intermediate; Figure 3b). Finally, Lily^b, an introductory MB instructor, discussed both phosphate transfer and common intermediate. However, Lily^b also emphasized the importance of equilibrium and the concentrations of reaction sequence components in driving reactions forward (code: equilibrium; Figure 3c).

Instructors Who Explained Both Energy Release and Transfer.

We were particularly interested in the nine instructors who leveraged both mechanisms and whether there was a difference in how or when they discussed each mechanism. In taking a closer look at the excerpts in which these codes emerged, we found that five of these instructors explained energy transfer in direct response to how ATP drives unfavorable processes, with their explanations for energy release occurring at different points or in different contexts during the interview. For example, Lily^b explained how energy is transferred, but they also talked about how it is common for biologists to discuss the energy release mechanism, saying: “… a lot of times […] biologists will just assume that well, okay. So there was this ATP, was broken, […] and usually they’ll say it was hydrolyzed. And that has 30.5 kJ/mol, right? And then I can do this other reaction because I’ve already paid, but without realizing that there’s no mechanistic coupling and that, that wouldn’t work” (codes: referencing disciplines and ATP hydrolysis).

The other four instructors (all biologists) initially explained energy release in direct response to “how do you think about the mechanism by which ATP drives unfavorable processes?”, but directly following this explanation, they leveraged the ideas of phosphate transfer and/or common intermediate to explain the mechanism of energy transfer. For example, Monstera^b, an introductory MB instructor, used ATP hydrolysis and bond energy when explaining how ATP drives unfavorable processes: “[…] we talked about a high energy phosphate bond in the context of reactants and products. And so that’s what we mean by when we say […] ATP plus water (code: ATP hydrolysis) is going to have more […] energy that is available” (code: bond energy). They continued talking about ATP hydrolysis but eventually said, “instead of having the phosphate just go to free phosphate, let’s donate it (code: phosphate transfer) to another molecule and have it become a phosphorylated intermediate” (code: common intermediate), thereby shifting their explanation to that of energy transfer.

Like Monstera^b, three other biology instructors who explained energy release via bond formation included ideas about phosphate transfer and/or common intermediate later during their interview, even though they initially used explanatory black boxes for how energy is transferred (Figure 4). While we recognize our small N value, we also found it interesting (and perhaps not surprising) that all three biochemists explained the energy transfer mechanism, with their ideas about energy release surfacing in other contexts (e.g., when referencing how other disciplines discuss the role of ATP). Although the majority (N = 9) of instructors discussed both mechanisms at some point, five of the instructors (three chemists and two biologists) only explained energy release and, therefore, continuously black boxed the mechanism of energy transfer (i.e., they had zero codes for both phosphate transfer and common intermediate; Figure 4). We do not assume whether they recognize these ideas as components of the mechanism of energy transfer; however, as neither came up in their interviews, even if they hold a mechanistic understanding, they did not see it as relevant to our discussion. In other words, only ideas related to energy release were activated, and not any other conceptualizations that these instructors might hold regarding this phenomenon.

While this analysis provided insights into the mechanisms that instructors embrace (transfer, release, or both), it also surfaced the unprompted, negative experiences that instructors expressed regarding teaching ATP (RQ3).

Research Question 3 – What Teaching Experiences Did MB Instructors Share Regarding ATP in Their Course(s)?

As evidenced in our interview protocol, we did not intend to elicit themes related to affect; however, several instructors shared their thoughts regarding pedagogical experiences with ATP (as evidenced by the frequent occurrence [n = 61] of the teaching/learning code). That is, rather than just explaining their current instructional practices, they also shared how they, as instructors, and/or their students, as learners, experience these instructional practices. Investigating this research question is warranted, 1) because it is likely that these experiences impact pedagogical/instructional approaches, and 2) because of its emergence despite minimal prompting. In this section, we highlight experiences shared by three introductory MB instructors who shifted from an explanation of energy release to that of energy transfer (as seen in RQ2). We chose these three participants because they teach the introductory course that most often includes a discussion of the role of ATP, and they provided particularly rich discussions about their experiences with instruction on this topic. Because the teaching/learning code captured a broad range of experiences rather than more specified affective constructs, we share longer excerpts (Figures 5, 6, and 7) to exemplify the dissatisfaction that these instructors expressed while reflecting on their teaching of ATP.

Jade ^b.

Consider the excerpt from Jade^b in Figure 5, whose response to the initial question about ATP providing energy put them in the energy release theme, because they leveraged ATP hydrolysis. However, they showed concern and vulnerability at several points in this excerpt that should not be ignored. Jade^b began by saying they’re “nervous” to explain the mechanism, because they know “[they] don’t get it all right.” While this was not further explored by the interviewer, there is a distinct possibility that this is a result of interactions with chemistry faculty and recent research publications (Dreyfus et al., 2014; Green et al., 2021; VandenPlas et al., 2021) that have focused on the bond breaking/energy release misconception often associated with instruction about ATP. In fact, some MB instructors noted that they focus specifically on what their chemist colleagues tell them (see excerpts in Supplementary Figure S3, captured by the code referencing disciplines).

Following this admission, interviewer K.N. attempted to mitigate Jade’s^b nerves/concerns by restating that this interview is not a test, and Jade^b proceeded to note two reasons for why they do not explain anything beyond “pairing the hydrolysis of ATP with unfavorable reactions.” First, they said, “I’m not sure it matters…that students understand that next level in order to understand the bigger ideas,” and they went on to say that “at some point, we had to assume that they had this knowledge when they get into cell and MB, that they understand…that there’s an intermediate and the intermediate is less stable than the final thing…” (code: common intermediate and teaching/learning). After K.N. validated some of Jade’s^b thoughts ([Editorializing] in Figure 5), they continued to discuss a third, more personal, reason for not going to a deeper level; Jade^b said, “the course seems sort of overwhelming and exhausting to me. I think if I understood it better, I would be able to better explain why molecular reactions happen in general” (code: teaching/learning). In just this excerpt, Jade^b shared negative experiences (including nerves, lack of confidence in content, and concern that this content is irrelevant or repetitive), that we interpret as dissatisfaction regarding their current teaching practices about the role of ATP in their course and/or dissatisfaction in the expectations surrounding teaching ATP. That is, regardless of the approach they take (energy release or energy transfer), there is some discomfort in the current pedagogy.

Monstera ^b.

Recall Monstera^b, an introductory MB instructor, who we saw in RQ2 discuss both ATP hydrolysis and how energy is released, as well as a follow-up explanation for energy transfer. In this subsequent discussion, Monstera^b revealed: 1) their understanding of the mechanism by which energy is transferred by leveraging both phosphate transfer and common intermediate, and 2) the challenges they experienced in teaching the role of ATP to students. Consider the excerpt in Figure 6.

Like Jade^b, Monstera^b initially focused their discussion on ATP hydrolysis and the bond energies associated with the reaction components to explain the mechanism for energy release. In probing further, we learned that while those ideas were the focus of instruction, Monstera^b specifically noted that they “don’t really get into how we couple those reactions…I would hope that that would be something that we can incorporate into a chemistry curriculum…” (code: referencing disciplines, and explicit exclusion of a mechanism). This quote provides evidence that Monstera^b intentionally does not address the mechanism of energy transfer in their course. Following this admission, Monstera^b specifically identified the mechanistic step of forming a phosphorylated intermediate (codes: common intermediate and phosphate transfer), suggesting that this be incorporated into a chemistry curriculum, because when they tried to do it in their course, they “had neither the expertise nor the time to do it.” Here, similar to Jade^b, Monstera^b expressed lack of confidence in the content and concern for course constraints (time). Finally, though less specific to ATP, Monstera^b referenced the course textbook and how the chapter on bioenergetics is “a sticky mess” and “it’s hard” for the students, further highlighting the negative experiences that they encounter as an instructor teaching these ideas in introductory MB.

Ficus ^b.

Lastly, Ficus^b, who primarily taught introductory MB but also biochemistry, discussed the challenges of deciding what level to go to when teaching these ideas in their course(s). Ficus^b only explained the mechanism of energy release, and their excerpts are shown in Figure 2 under RQ2. In addition to these excerpts, Ficus^b shared the challenges of teaching these ideas to students. Consider the excerpt in Figure 7.

In this excerpt, Ficus^b shared their concerns with teaching the mechanism by which ATP hydrolysis releases energy in an introductory MB course. At no point did they bring in ideas about phosphate transfer or common intermediate. When we first asked how they explain the mechanism of how ATP works, Ficus^b said they tried teaching the mechanism (of how ATP hydrolysis releases energy, which they later note) in the past “with very limited success…caused way more problems than it should have”. Because of this, they actively avoid teaching these ideas, because they “found it to be not productive”. Ficus^b, in our interpretation, expressed frustration and dissatisfaction when reflecting on their past experiences with teaching ATP by explaining the mechanism of energy release to introductory MB students. In their words, their efforts resulted in “way more problems”, and were “not productive”, and created “mass confusion” (code: teaching/learning).

Our use of a semistructured interview protocol allowed for these rich discussions to emerge. While their emergence is perhaps not surprising given the purpose of the interviews, the findings here show that minimal prompting activated personal and meaningful experiences that have shaped how these instructors think and teach about ATP. We provide shorter excerpts from additional instructors who shared negative experiences in Supplementary S2 (Supplementary Table S1).

DISCUSSION

In this study, we set out to understand how a range of chemistry, biology, and biochemistry instructors think about the mechanism by which ATP drives unfavorable processes in relation to their discipline and course(s). The interviews revealed the range of ideas instructors think about when considering the roles of ATP (RQ1). From the analysis in RQ2, we found that the instructors primarily focused on two mechanisms when explaining how ATP drives energetically unfavorable processes: 1) energy transfer through the phosphorylation of a common intermediate and 2) energy release based on the breaking of weaker bonds and the forming of stronger bonds via ATP hydrolysis. Between the two, we found that most faculty leveraged the energy release mechanism (N = 14); however, nine of these instructors leveraged both mechanisms. In this case, some instructors described their use of the energy release mechanism when teaching about the roles of ATP but indicated understanding of the energy transfer mechanism as disciplinary experts (e.g., Monstera^b and Jade^b). Other instructors, like Lily^b for example, focused on the mechanism of energy transfer in explaining the role of ATP, but explicitly brought up how certain disciplines might only explain the mechanism of energy release.

Our interviews showed that current instructional practices about ATP in biology courses appear to be dissatisfying and frustrating for instructors (RQ3). Within the three longer excerpts, we also noticed some comments which suggest that the instructors did not see certain aspects as relevant to a general understanding of biological systems (e.g., Jade^b said “I’m not sure it matters…”), perhaps speaking to the lack of empowerment of instructors. Further, it is likely that these instructors have conveyed this assumption to students. Some instructors shared the frustrations they experienced when teaching the energy-release mechanism, especially Ficus^b who said that teaching this mechanism “caused way more problems than it should have”. These results highlight how current approaches are dissatisfying (evidenced by the negative experiences described by instructors). Additionally, as noted by Lily^b, this approach can be a misleading way to discuss the role of ATP: “… usually they’ll say it was hydrolyzed. And that has 30.5 kilojoules per mole, right? And then I can do this other reaction because I’ve already paid, but without realizing that there’s no mechanistic coupling and that, that wouldn’t work”.

Energy Release: A Misleading Mechanism

While both the energy transfer and energy release mechanisms were explained accurately by the instructors (i.e., we did not note any misconceptions), the energy release mechanism explains a different phenomenon (how/why ATP hydrolysis releases energy in an isolated system) and is inconsistent with the actual molecular mechanistic events when ATP is involved in reaction coupling. Even a canonical discussion of ATP hydrolysis and bond energy does not address how the energy released from ATP hydrolysis drives the unfavorable process. While both energy release and energy transfer are important and related concepts, the two mechanisms often appear to be conflated with one another.

Lipmann (1941) used the ATP hydrolysis reactions to calculate bond energies, but his experimental design did not suggest that the hydrolysis of ATP was the biological mechanism for the transfer of energy. We are not the first to identify this issue in the discussion of the biological mechanisms of ATP – over half a century ago Banks and Vernon (1970) noted that “simple thermodynamic parameters are irrelevant in discussing whole organisms: these must be understood in kinetic and mechanistic terms”. Later in his career, Lipmann himself remarked that defining ATP as a phosphoryl group donor (i.e., the energy transfer mechanism) “could lead to a better understanding of such sequences of energy transfer where we are pretty much in the dark about events subsequent to an initial ATP-involving step” (Lipmann, 1960). The conflation of the energy release and energy transfer mechanisms in biology has led to much confusion about what is actually happening in biological systems and how the materials should be taught, so it is not surprising that this mechanism was discussed by nearly all of the instructors (n = 14). However, no matter how well we improve the accuracy by which we discuss the energy release associated with ATP hydrolysis (by including an explicit discussion of the role of water and bond energies), this disconnect will not be addressed.

We do not believe anyone should be blamed for the conflation of these mechanisms, as our interviews provide evidence that biology instructors have a strong desire to align their courses with chemistry (see Supplementary Figure S3 excerpts from biologists) by communicating how ATP works in a way that mitigates the misconception that breaking bonds releases energy. Not only did Lipmann use bond energies to explain why ATP is a metabolically useful molecule (Lipmann, 1941), but bond energies are an important idea in chemistry and physics. Thus, it makes sense to focus on the importance of bond energies when discussing why ATP hydrolysis is a thermodynamically favorable reaction. However, if we focus solely on ATP hydrolysis, we miss out on the biological mechanisms by which ATP actually drives phenomena such as unfavorable chemical reactions, the mechanism which all of the biochemists and most of the biologists explained at some point in their interviews (suggesting its importance in the discipline).

The Need for Additional Research on “Energy Transfer”

In our introduction, we noted the literature base in which researchers have explored the issue of ATP’s role in biological systems; nearly all this work (based in physics and chemistry) has focused on ATP hydrolysis and the energy release mechanism. At the same time, no biology education research (to our knowledge) has focused on energy transfer mechanisms. That is, the ways in which ATP “provides” energy via transfer of (usually) its terminal phosphoryl group to the starting material, thereby activating a substrate or reactant (increasing its energy/reactivity), rather than through an adjacent reaction with water. Additionally, other mechanisms, such as the alteration in protein structure when ATP is bound, might also be more accessible if the current focus on ATP hydrolysis is removed. If we can understand how to incorporate these mechanisms into our undergraduate courses, specifically introductory biology, we may alleviate some of the negative experiences reported by instructors, while simultaneously helping students to be better equipped to make mechanistic predictions regarding the role of ATP in other biological processes. As a result, given the association between ATP and energy, this may help students develop and understand ideas about energy that are compatible across disciplines.

Research in this area is still in its infancy. We hope this study encourages the education research community to think deeply about the potential utility of the energy transfer mechanism (which we discuss below), as many important questions remain unanswered. For example, how do students consider both the energy release and energy transfer mechanism? What level of explanatory depth is necessary and most appropriate for discussing these ideas in the disciplines (or different course levels)? How do students with an understanding of the energy transfer mechanism think about energy across disciplines? How does this mechanism contribute to other important ideas in biology like structure/function? The exploration of such questions is critical to the development of instructional materials that might better support the instructors who are tasked with teaching these ideas.

Suggestions for Instruction

Here we provide pedagogical suggestions that aim to better support instructors as they approach this topic in their course(s). Our findings emphasize that current approaches can lead to frustration for some instructors, and, if including discussions about energy release, misleading ideas for students. Further, most of the biology instructors (n = 4), all of the biochemistry instructors (n = 3), and only one chemistry instructor discussed phosphate transfer and/or common intermediates to explain the mechanism of energy transfer, suggesting its relevance in biology and biochemistry. Based on literature providing evidence for the usefulness and importance of mechanistic reasoning in education as outlined in the Introduction, we suggest the incorporation of an energy transfer mechanism and/or explicit recognition of explanatory black boxes, as proposed by Haskel-Ittah (2023). Recall that an explanatory black box is a “unit” within a mechanistic explanation that remains unexplained – that is, the entities and interactions which give rise to that unit are not identified or discussed, such that a gap is created between steps within the mechanistic explanation (Haskel-Ittah, 2023). Explanatory black boxes, while often referred to as “hand-wavy” explanations that lack evidence of deeper knowledge, provide great utility in biology education because of the complexity of biological systems and the diversity of mechanisms involved – it is unreasonable, impractical, and time-consuming to provide a fully mechanistic explanation for every phenomenon without any black boxes. Thus, deciding the relevant mechanism on which to focus becomes just as imperative as it is challenging. To add to the challenge, this decision must also be paired with explicit recognition of existing explanatory black boxes so as to avoid what has been termed an “illusion of explanatory depth” (Haskel-Ittah, 2023); that is, the sense that one understands complex phenomena deeply, when that may not be the case. We believe this approach will alleviate the pressures felt by instructors to explain topics that they feel “nervous” about or that they find irrelevant to an understanding of biological systems (i.e., the energy release mechanism).

To illustrate how this could be done, we discuss the unfavorable, ATP-dependent process of the formation of glutamine (Figure 8). Typically, this process is shown in such a way that the mechanism by which ATP drives the reaction is presented as an explanatory black box, similar to how we presented it in words to participants in the interviews. Only the input and output are provided, and the mechanistic steps are missing (“phenomenon-based sketch”). Without a description/model of the entities that interact or link together to contribute to the outcome (i.e., the formation of glutamine), we can identify that ATP is involved, but not how it is involved. Further, without explicitly recognizing this as an explanatory black box, it is likely (and reasonable) that the learner will apply the mechanism that they know, the energy-release mechanism, to explain how this unfavorable reaction occurs. However, such an application of the energy-release mechanism does not provide a physical mechanism for how the energy is transferred from the hydrolysis to the reaction between glutamate and ammonium. We predict that by explicitly stating there is more going on at these reaction arrows (i.e., additional relevant entities are not shown for simplification purposes), we both: 1) better prepare the student for future learning about this topic, and 2) mitigate unwanted Dunning-Kruger effects (Dunning, 2011) associated with the overestimation of one’s understanding of a phenomenon (i.e., avoid the illusion of explanatory depth).

While explicitly recognizing this explanatory black box can prepare students for future learning, it still does not address the multiple mechanistic roles played by ATP, common processes that can (and perhaps should) be emphasized given the biological importance of this molecule. We posit that unpacking this black box by leveraging the energy-transfer mechanism would support students’ understanding of the mechanistic role of ATP, and thus, their ability to explain or predict other biological processes involving ATP or energy transfer in these complex systems.

Using the energy-transfer mechanism outlined by both biology and biochemistry interviewees, we can describe the transfer of a phosphoryl group from ATP to glutamate, forming the phosphorylated glutamate. This more reactive common intermediate can then react with ammonium, releasing inorganic phosphate and producing glutamine. Broadly, a similar approach could be applied for phenomena in which phosphoryl transfer does not occur, like the noncovalent interactions formed with ATP binding to an enzyme. This physical interaction, through either noncovalent or covalent forces, is key to explaining the mechanism of energy transfer. This approach is also in better alignment with the actual mechanisms by which the biological phenomenon occurs, as glutamine formation does not involve the hydrolysis of ATP and instead occurs through the formation of a phosphorylated intermediate. Further, we propose investigating the role of prior knowledge in understanding and explaining these ideas, specifically by focusing on ideas about reactivity, which are often introduced in chemistry courses. Using this approach unpacks how ATP provides energy; however, additional explanatory black boxes which could be further unpacked still remain, for example, how and why does ATP react with glutamate? or how does the phosphoryl group change the reactivity of glutamate? These additional black boxes highlight the complexity of a relatively simple biological phenomenon, and therefore, the importance of making explanatory black boxes explicit, so as to best prepare students for future learning. This approach is our best recommendation based on the evidence uncovered in this study, our understanding of the roles of ATP, and the literature supporting mechanistic reasoning and its role in undergraduate science education; however, future work should investigate the effectiveness of these specific approaches. Based on this future work, we will refine these approaches to provide more effective materials to support students’ learning of the mechanism of a key role by which ATP works in cells.

LIMITATIONS

Our goal in sharing much of our qualitative data was to highlight the authentic thoughts of each participant; however, we recognize our biases, regardless of our attempts to mitigate these biases as outlined in our positionality statement. We urge our readers to think deeply about their own interpretations of these interview excerpts, as they may differ slightly from ours. We are grateful to the instructors for trusting us with their thoughts and have sincerely aimed to share this data appropriately and carefully.

Our focus with this work has been on chemical reactions, but there are other cellular events in which ATP acts as a regulator, chaperone, signaling molecule, hydrotrope (Patel et al., 2017), etc. These additional roles are critical in biology, as well as events leading to the synthesis of ATP and the maintenance of high (mM) and steady intracellular ATP concentrations. We plan to investigate these areas in future projects and encourage other experts to do the same. This work reflects a focus on chemical reactions in biological systems because of our concern with developing interdisciplinary teaching and learning techniques that help students link ideas between their undergraduate chemistry and biology courses.

Several of the instructors we interviewed are involved in education research, and perhaps a different group of instructors would include different ideas or different combinations of ideas regarding both their teaching and disciplinary understandings regarding ATP.

Lastly, we used the teaching/learning code to represent the experiences instructors shared with us during their interviews. This code captured a wide range of experiences, which we did not categorize into subcodes; however, additional investigation should be done to characterize how instructors feel about their current, past, or proposed teaching practices regarding this critical topic so as to best support them and, in turn, their students.

ACKNOWLEDGMENTS

REFERENCES