Schizophrenia Research Forum - A Catalyst for Creative Thinking

Matsuda Y, Marzo A, Otani S. The presence of background dopamine signal converts long-term synaptic depression to potentiation in rat prefrontal cortex. J Neurosci. 2006 May 3 ; 26(18):4803-10. Pubmed Abstract

Comments on News and Primary Papers
Comment by:  Andreas Meyer-Lindenberg
Submitted 15 May 2006
Posted 15 May 2006
  I recommend the Primary Papers

I think this is an interesting paper, as it shows that alterations in tonic dopaminergic stimulation can result in a pronounced and qualitative switch (LTD to LTP) in the behavior of prefrontal neurons. Although the concept of tonic versus phasic dopaminergic stimulation has been adopted widely by the schizophrenia research community, the majority of the preclinical work has focused on acute changes in dopamine concentration and on subcortical structures, especially the nucleus accumbens, and from my perspective as a clinical researcher, it is welcome to see some data that extend to prefrontal cortex and longer timescales, although it must be emphasized that this paper concerns results from rats, in slices in vitro, using tetanic stimulation, and that the pretreatment with dopamine lasted for 40 minutes only. With these caveats, it is exciting to see that pretreatment with dopamine after what the authors presume is a 4-hour period of neurotransmitter depletion during slice preparation produces LTP after a weak tetanic stimulus, compared to LTD that the same stimulus evoked without dopamine priming. Since LTD arose under conditions of relative dopamine depletion, which might reflect, at least in directionality, the situation in schizophrenia, these data suggest that functionally impairing qualitative changes in a neuronal response in prefrontal cortex of relevance for working memory function could result from quantitative reductions in extracellular (tonic) dopamine content. It is also of interest that the authors demonstrate that the LTP requires concurrent stimulation of metabotropic glutamate receptors, suggesting a mechanism by which widely studied risk genes for schizophrenia such as COMT and GRM3 could interact in impairing prefrontal cortex function.

View all comments by Andreas Meyer-LindenbergComment by:  Patricia Estani
Submitted 3 June 2006
Posted 3 June 2006
  I recommend the Primary PapersComment by:  Terry Goldberg
Submitted 20 June 2006
Posted 20 June 2006

Matsuda et al. demonstrate that priming D1 and D2 receptors may induce LTP; otherwise, LTD develops. To elaborate, a weak tetanic stimulation and dopamine stimulation produces LTD. However, if dopamine is perfused for 12 to 40 minutes at D1 and D2 receptors and a tetanic stimulus is provided, LTP, a form of cellular learning associated with memory, develops. This study has potentially important implications for understanding the cause of prefrontally based failures in information processing in schizophrenia. It gives additional weight to arguments that reduced dopaminergic tone at the cortical level is responsible for at least some of the cognitive problems associated with the disorder.

It also helps make sense out of some otherwise anomalous data in the literature. For instance, in manipulations of several tests of purported attentional control and vigilance problems, findings appeared more consistent with difficulties in constructing a representation than with attention per se in target detection (e.g., Elvevag et al., 2000; Fuller et al., 2005).

One thing that I would certainly give an eyetooth to know is how the authors view their work in light of findings by Seamans and Goldman-Rakic on differences in the consequences of stimulation of D1 and D2 receptors (simplistically, that D1 activation promotes task-relevant information, while D2 stimulation may produce task-irrelevant information processing experienced as interference).

Caveat Emptor: I don’t have the expertise to comment on the slice preparation methodology.


Elvevag B, Weinberger DR, Suter JC, Goldberg TE. Continuous performance test and schizophrenia: a test of stimulus response compatibility, working memory, response readiness, or none of the above? Am J Psychioatry 2000; 157:772-780. Abstract

Fuller RL, Luck SJ, McMahon RP, Gold JM. Working memory consolidation is abnormally slow in schizophrenia. J Abnorm Psychol 2005; 114:279-290. Abstract

View all comments by Terry GoldbergComment by:  Satoru Otani
Submitted 22 July 2006
Posted 24 July 2006

In his June 20 comment, Dr. Goldberg raised an important question concerning our paper: how our results, showing the necessity of D1+D2 receptor coactivation for prefrontal LTP induction and priming, fit into the scheme proposed by Seamans et al., 2001, that is, the differential roles played by D1 and D2 receptors for prefrontal cortex (PFC) cognitive processes.

I think I have to first point out that the dependency of PFC long-term potentiation (LTP) induction (let alone "priming" now) on DA receptor subtypes may vary among subpopulations of PFC synapses. In ventral hippocampus (HC)-PFC synapses, LTP induction requires D1 but not D2 receptors (Gurden et al., 2000). This in vivo study fits with the idea that HC-PFC projection and its D1 receptor-mediated modulation are critical in spatial information processing (working memory) and encoding of this information. However, recent in vivo results of Yukiori Goto at the University of Pittsburgh (personal communication, but see Goto and Grace, 2005) indicate that LTP induction in cortico-cortical synapses in the PFC may be dependent on both D1 and D2 receptors, similar to our case. Dr. Goto found that while synaptic potentiation in the HC-PFC synapses indeed depends only on D1 receptors, potentiation in cortico-cortical synapses, stimulated by the electrode inserted in the superficial layer of the PFC as in our preparation, depends on the activation of both D1 and D2 receptors. Thus, it appears that DA receptor dependency of LTP induction differs between the HC projection input and the cortico-cortical input—the former dependent only on D1 receptors and the latter on D1+D2 receptors.

How significant this difference might be functionally is, of course, still an issue for speculation. It seems clear that working memory input from the HC (and strengthening of this input), which may depend only on D1 receptors, are critical for PFC cognitive function. But also, other cortical inputs, which are not necessarily related to the attention-driven working memory, may be as critical for the formation and achievement of goal-directed behavior, and strengthening of these cortical inputs may depend on D1+D2 receptors. Incidentally, Dr Goto also showed that the organization of a planned behavior tested in a modified radial-arm maze task requires not only intact HC-PFC connection but also the activation of both D1 and D2 receptors within the PFC (Goto and Grace, 2005). We are tempted to suggest that neuronal traces within the PFC necessary for the generation of goal-directed behavior may be heterogeneous both in their input origin and in their formation mechanism.


Seamans JK, Gorelova N, Durstewitz D, Yang CR (2001) Bidirectional dopamine modulation of GABAergic inhibition in prefrontal cortical pyramidal neurons. J Neurosci 21, 3628-3638. Abstract

Gurden H, Takita M, Jay TM. Essential role of D1 but not D2receptors in the NMDA receptor-dependent long-term potentiation at hippocampal-prefrontal cortex synapses in vivo. J Neurosci. 2000 Nov 15;20(22):RC106. Abstract

Goto Y, Grace AA (2005) Retrospective and prospective memory processing in the hippocampal—prefrontal cortical network. Soc Neurosci Abstr 413.3.

View all comments by Satoru OtaniComment by:  Jeremy Seamans
Submitted 26 July 2006
Posted 27 July 2006

Drs. Goldberg and Otani raise some excellent points in their comments on the Matsuda et al. paper. As Dr. Otani alluded to in his latest comment, it is useful to define exactly what is being modulated under different experimental conditions and how this all relates to prefrontal cortex (PFC) function in general.

Dr Otani’s studies investigate synaptic plasticity induced by tetanic stimulation and how this process is modulated by tonic dopamine (DA). Long-term potentiation/long-term depression (LTP/LTD) induced by tetanic stimulation has provided us with perhaps the best model of the cellular basis of long-term memory and has been proposed to underlie, among other things, various aspects of long-term spatial memory and declarative memory. LTP is a long-lasting, passive, associational memory mechanism, unlike working memory that is transient in nature, relies on active processing and is not associational. Therefore, in PFC, it would be highly unlikely that LTP/LTD is the neural mechanism of working memory. However, to solve a working memory problem, one must manipulate newly acquired information within a certain context or based on a pre-learned rule. Perhaps the best example of how these processes relate can be found in White and Wise, 1999, and Wallis et al., 2001, who investigated the activation of PFC neurons in situations where two different abstract rules could be applied. PFC neurons showed different degrees of activation during a delay period depending on the preference of the neuron for a specific task rule. Therefore, stable long-standing rules regulate how strongly a cell in PFC exhibits short-term memory related activity. These rules were learned over time and were stable. LTP/LTD are as good mechanisms as any for their cellular basis. This implies that LTP/LTD-like mechanisms influenced the manner in which PFC neurons exhibited transient working memory related activity. Therefore, as shown by Matsuda et al. and suggested by others (e.g., Lisman and Grace, 2005), a long-term memory mechanism, perhaps involved in the formation of stable rules, is subjected to modulation by tonic and phasic DA. This long-term memory in turn regulates the online active processing of information in working memory.

In contrast, many investigators have proposed that DA is also able to directly modulate working memory related activity. As noted by Dr. Goldberg, in addition, we have suggested that the mode of modulation is different for D1 and D2 receptors in PFC. Like task rules, DA appears to modify the strength of delay-period activity (e.g., Sawaguchi, 2001). Furthermore, the modulation of synaptic currents by DA, especially via D1 receptors, is very long-lasting and has been termed “late potentiation” and in fact shares aspects of the late phase of LTP (Huang and Kandel, 1995). However, unlike stable task rules, DA levels can change quickly and dynamically, and as a result, delay-period activity may be increased or decreased depending on the level of DA and the differential activation of D1 and D2 receptors, even if the same task rule is being implemented. The dynamic modulation of DA levels depends on a variety of factors such as intrinsic motivation, stress, and even the strength of the active memory trace (Phillips et al., 2004).

Therefore, DA modulates LTP/LTD, which in turn may be involved in the rule-dependent modification of delay-period activity. DA can also directly modulate the ionic currents involved in actually generating delay-period activity. Although this modulation can be long-lasting, DA levels and activation of different DA receptors change dynamically and the mode of modulation could continuously vary based on a variety of intrinsic and task-dependent variables.

Perhaps one implication of all this for schizophrenia would be that dysfunction of DA-dependent modulation of LTP/LTD would lead to an inability to accurately store or implement the appropriate rule for a given situation. In contrast, dysfunction of the direct DA modulation of ionic and synaptic currents could lead to more immediate issues such as distractability or pathologically focused processing of information within working memory (Seamans et al., 2001; Seamans and Yang, 2004).


Huang YY, Kandel ER. D1/D5 receptor agonists induce a protein synthesis-dependent late potentiation in the CA1 region of the hippocampus. Proc Natl Acad Sci U S A. 1995 Mar 28;92(7):2446-50. Abstract

Lisman JE, Grace AA. The hippocampal-VTA loop: controlling the entry of information into long-term memory. Neuron. 2005 Jun 2;46(5):703-13. Review. Abstract

Phillips AG, Ahn S, Floresco SB. Magnitude of dopamine release in medial prefrontal cortex predicts accuracy of memory on a delayed response task. J Neurosci. 2004 Jan 14;24(2):547-53. Abstract

Sawaguchi T. The effects of dopamine and its antagonists on directional delay-period activity of prefrontal neurons in monkeys during an oculomotor delayed-response task. Neurosci Res. 2001 Oct;41(2):115-28. Abstract

Seamans JK, Gorelova N, Durstewitz D, Yang CR. Bidirectional dopamine modulation of GABAergic inhibition in prefrontal cortical pyramidal neurons. J Neurosci. 2001 May 15;21(10):3628-38. Abstract

Seamans JK, Yang CR. The principal features and mechanisms of dopamine modulation in the prefrontal cortex. Prog Neurobiol. 2004 Sep;74(1):1-58. Review. Erratum in: Prog Neurobiol. 2004 Dec;74(5):321. Abstract

Wallis JD, Anderson KC, Miller EK. Single neurons in prefrontal cortex encode abstract rules. Nature. 2001 Jun 21;411(6840):953-6. Abstract

White IM, Wise SP. Rule-dependent neuronal activity in the prefrontal cortex. Exp Brain Res. 1999 Jun;126(3):315-35. Abstract

View all comments by Jeremy Seamans