Category: 3. Business

In this Help Net Security interview, Michael Yaroshefsky, CEO at MCP Manager, discusses how Model Context Protocol’s (MCP) trust model creates security gaps that many teams overlook and why MCP must not be treated like a standard API. He explains how misunderstandings about MCP’s runtime behavior, governance, and identity requirements can create exposure. With MCP usage expanding across organizations, well-defined controls and a correct understanding of the protocol become necessary.

What aspects of MCP’s trust model are most misunderstood right now, and can you share a real example where implementers made incorrect assumptions?

Many people hold an erroneous (and dangerous) assumption that communication between MCP servers and clients is essentially the same as API-based transactions. However, MCP and APIs are incredibly different, especially when it comes to your security posture. It’s dangerous to think otherwise. 

APIs generally don’t cause arbitrary, untrusted code to run in sensitive environments. MCP does though, which means you need a completely different security model. LLMs treat text as instructions, they follow whatever you feed them. MCP servers inject text into that execution text. For example, “what tools exist? What are the descriptions for these tools?” 

That text can influence LLM behavior. Further, unlike APIs where you can use a specific API version, you can’t review and pin trusted versions within an MCP environment. Upon each connection, your MCP client will receive the latest published metadata provided by the MCP server. In other words, MCP provides runtime-provided text that you have no way to inspect. While an MCP server may seem benign upon initial connection, there’s the latent possibility for a trusted MCP server to inject malicious context in the future. That would be called a rug pull. These risks are unique to MCP, and they require specialized solutions that ordinary API security frameworks cannot provide. 

Security professionals might also erroneously assume that they can trust all clients registering with their MCP servers, this is why the MCP spec is updating. MCP builders will have to update their code to receive the additional client identification metadata, as dynamic client registration and OAuth alone are not always enough. 

Another trust model that is misunderstood is when MCP users confuse vendor reputation with architectural trustworthiness. Ever since the MCP spec began supporting streamable HTTP transport, reputable SaaS vendors could easily publish MCP servers that users can then run by any local- or cloud-based MCP client. However, teams shouldn’t assume that first-party servers from reputable companies are immune to security vulnerabilities. 

For example, researchers have also uncovered prompt injection vulnerabilities with GitHub’s MCP server and Atlassian’s servers in May and June of this year. There was also a report about Microsoft Copilot still being at risk of prompt injection as well. So, you can’t assume that these servers are all safe.

Lastly, and most importantly, MCP is a protocol (not a product). And protocols don’t offer a built-in “trust guarantee.” Ultimately, the protocol only describes how servers and clients communicate through a unified language. MCP does not solve authentication and identity management, enterprise operations (e.g., audit trails, observability, compliance) and infrastructure (e.g., hosting, error handling, rate limiting). 

Organizations are beginning to deploy large numbers of MCP servers internally. What governance blind spots appear when MCP becomes a widespread integration fabric, and can you describe a case where poor governance created operational or security issues?

Organizations often lack centralized MCP observability and controls, leaving more room for vulnerabilities to emerge outside the purview of security team members. Many organizations don’t even have an internal MCP registry, which is table stakes for setting up processes to approve and govern MCP servers. 

When companies don’t have processes to approve and monitor MCP servers, shadow MCP and server sprawl both happen. With shadow MCP, employees introduce servers that IT knows nothing about (and wouldn’t approve). IT/security teams also can’t monitor that server’s security in the long run (e.g., if a server starts out fine but becomes vulnerable later on, they’d never know someone internally was using that, even if they became aware of this server having a vulnerability). Server sprawl happens when duplicative, unnecessary, or unused MCP servers create an ever-expanding attack surface.

MCP gateways allow companies to have an internal registry, which mitigates both shadow MCP and server sprawl. Internal registries make it clear to employees how to get approvals, allowing IT teams to provision tools and provision servers to teams.

We’ve onboarded a large number of teams that want to create MCP gateways after having poor governance wreck havoc in their organization. I’ve seen security leaders who felt burnt after teams deployed MCP servers locally without sandboxing, or using insecure token storage, access control, and scoping practices. Local MCP servers are especially dangerous, because they may have access to sensitive on-device credentials or files, there could be bearer or API tokens in an MCP.json file (which is concerning because they’re production-access tokens sitting on a machine). So, any vulnerability that can read files could suck those up and send them somewhere nefarious. 

These are the kinds of issues that security teams either encounter or foresee that causes them to seek an MCP governance solution. Because ultimately, poor governance gives rise to inconsistency of deployment methods, auth processes, and identity management, which can introduce further, wide-ranging risks, and make your MCP ecosystem even more difficult to provision, observe, and fortify.

As more models gain the ability to call MCP tools, the risk of unauthorized agents or spoofed contexts grows. What steps should organizations take to verify that both the MCP server and the invoking model are authentic, and what protections are still missing from the specification?

Firstly, organizations should create a review and approval process for adding all MCP clients and servers. This will help protect them from supply chain risks, and it can reduce the likelihood of team members inadvertently introducing malicious clients and servers into the organization. 

Security conscious organizations should also insist that all MCP servers use OAuth 2.1 with Proof Key for Code Exchange (PKCE) and harden their approach by ensuring that they use regularly rotated, finely-scoped, and securely stored tokens. OAuth is the recommended (but not required) auth flow in the MCP spec because other, more basic auth flows aren’t always time-scoped, which can give access for longer than any IT professional would want. It can be risky to use bearer tokens (instead of OAuth) because they’re often stored in plain text on a machine, which can be used for nefarious purposes if a local MCP server is compromised. 

Risks can also emerge from the names of tools within MCP servers. If tool names are too similar, the AI model can become confused and select the wrong tool. Malicious actors can exploit this in an attack vector known as Tool Impersonation or Tool Mimicry. The attacker simply adds a tool within their malicious server that tricks the AI into using it instead of a similarly named legitimate tool in another server you use. This can lead to data exfiltration, credential theft, data corruption, and other costly consequences. 

Implementing and mandating the use of an MCP gateway in your organization provides a solution to most of these risks, as it enables you to:

• Create and manage your organization’s server and client registry
• Standardize and ensure the robustness of all MCP auth flows
• Ensure proper token rotation
• Create allowlists and blocklists for MCP servers, tools, and clients
• Add namespaces to tools to assist the AI model in selecting the correct tool

Where do you think practitioners underestimate the operational effort required to run MCP securely? Is it observability, key management, server hardening, or something else, and what examples have you seen where teams were caught off guard?

Teams underestimate how much work it takes to implement strong access controls and permission boundaries when using MCP. In addition, the way that most enterprise companies handle identity management and authorization doesn’t always fit into what MCP requires for safe, secure, and scalable deployment. 

For example, the MCP specification relies upon processes like dynamic client registration (DCR) to register the MCP client with a server. Not all engineers are familiar with DCR because not all auth flows require it. But more importantly, enterprises don’t want anonymized auth flows or shared “service accounts” to access systems, data, applications, and other resources.

Enterprises we’ve worked with want MCP to plug into their existing identity management infrastructure. They also want real identities attached to both human users and AI agents, along with policies and control. 

However, implementing even the most basic level of identity and permissions management for MCP servers is a very heavy lift. In addition, there are a lot of flashy (and very dangerous) attack vectors with cool names that get more attention. Identity management, on the other hand, is complex, tricky, and continuously changing, as the capabilities of AI models, use cases for MCP, and the MCP specification itself all evolve. This is why identity management often gets overshadowed and overlooked.

I’ll sound like a broken record here but that’s where MCP gateways come in. When assessing an MCP gateway, ensure that it offers proper identity management. In addition, you’ll want a gateway that allows teams to provision gateways in such a way that requires each user accessing the gateway to use their own personal credentials for the MCP servers. This prevents the overuse or abuse of shared credentials or “bot” / “service” accounts that may provide too much access and not enough auditability.

What do you see as the most significant governance challenge as MCP adoption expands across industries, and which emerging best practice do you expect to become standard within the next year?

Regulatory compliance will become an increasingly important governance challenge as MCP adoption expands across industries and jurisdictions. Using MCP servers creates real risks around data security, protection, and privacy. 

If organizations don’t have strict, granular access controls and guardrails against sensitive data use and exfiltration, then they will face internal pressure to implement them, along with external pressure. Organizations may need to create or implement measures to comply with near-future legislation that specifically addresses the use of personal data, financial information, health records, and other highly regulated data by AI models. That will include safeguards to prevent data from being accessed by AI or ways to provide auditable logs of AI’s access and actions based on this information.

I think any security professional who has come into contact with MCP servers and begun to consider the implications for their organization will have concluded that an MCP gateway is a non-negotiable, essential tool they need to deploy, secure, manage, and monitor MCP servers. The best parallel may be how nearly all organizations have robust protections around corporate email, including strong multi-factor authentication requirements, anti-spam, anti-phishing, and audit logs. While you could use email without a platform offering these features, and for many years teams did, it’s an unnecessary risk, and nearly all organizations now use sophisticated email software with these capabilities. MCP governance platforms will become similarly ubiquitous as the ecosystem matures, so it’s just a question of when companies will adopt MCP governance capabilities.

In terms of other best practices, larger organizations will likely adopt policy-based access controls early on in their MCP adoption. Taking a policy-based approach is a more scalable, secure, and granular way to control access to resources and permissions that fits better with the unpredictable ways that agentic AI uses MCP servers and resources.

Lastly, many organizations are already deploying MCP servers as internal services, hosted in their own cloud. This shift towards managed MCP deployments will increase, and you’ll see fewer purely local or remote MCP deployments, at least within enterprises.

Australian state-owned utility CS Energy’s gas-fired 400MW Brigalow Peaking Power Plant in Queensland will come on line in late 2028, a year later than originally announced.

The delay comes as CS has formed a joint venture with Australian gas infrastructure firm APA to fund Brigalow, APA said on 1 December. APA will take a 80pc non-operated stake in the project for which it will building a connection to the gas grid, while CS will control the remaining 20pc and operate the facility.

Construction of the plant will take three years and will include the installation of 12 gas turbines, with the power plant now set to be commissioned at the end of 2028, a CS spokesperson said.

The project will cost about A$1bn ($650mn), an analyst with RBC Capital Markets said. The companies expect to complete an engineering design before June 2026, which will determine project costs.

CS recently announced a gas supply agreement with Australian gas company Senex Energy for up to 58.4PJ (1.56bn m³) over 10 years.

Queensland’s conservative Liberal National Party government included [A$479mn] (https://direct.argusmedia.com/newsandanalysis/article/2744404) in its 2025-26 state budget for the Brigalow peaking plant.

This investment is in line with the state’s five-year energy roadmap released in October, which outlines plans to keep coal-fired power plants operational until the late-2030s and mid-2040s and to introduce new gas-fired capacity.

Queensland’s electricity generation in the last 12 months consisted of 72pc black coal, 11pc solar and a 7pc share each for gas and wind, data from the Australian Energy Market Operator show. The state has the highest percentage of black coal generation in the national energy market, followed by New South Wales’ 68pc.

By Susannah Cornford

Stock Futures Little Changed as Traders Enter December

Expectations are running high for a strong December after a volatile November.

—-

China Private Gauge Signals Weaker Manufacturing Activity

Growth in China’s manufacturing production came to a halt as new orders nearly stalled last month despite a renewed rise in new orders from abroad.

—-

South Korea’s Export Growth Picked Up in November

South Korea’s exports rose at a stronger-than-expected pace in November, backed by brisk demand for semiconductors and a trade deal between Seoul and Washington.

—-

Australia’s Share-Market Operator Suffers Publishing Outage

Australia’s beleaguered stock exchange operator is investigating an outage that prevented dozens of companies from publishing investor updates at the start of the new trading week.

—-

Rate Hikes Wouldn’t Put Brakes on Japan’s Economy, BOJ’s Ueda Says

The Bank of Japan will thoroughly discuss the possibility of an interest-rate increase at its upcoming meeting, Gov. Kazuo Ueda said, stoking hopes for a resumption of monetary tightening this year.

—-

America’s Tariffs Jolted the Global Economy. Its AI Spending Is Helping Save It.

Economists predicted a global shock from President Trump’s tariffs, but some of them are now revising their global growth predictions upward.

—-

American Consumers Have Had It With High Car Prices

Shoppers are starting to draw the line on what they will pay for a new car, with some turning to used vehicles, taking on longer car loans and holding out for deals.

—-

Is America Heading for a Debt Crisis? Look Abroad for Answers

Politics and debt don’t mix well. Americans would be wise to look across the Atlantic to see how tough things can get.

—-

Since Trump’s Return, Bets on His Brand Have Soured

Stocks and cryptocurrencies tied to the president and his family have tumbled amid a broader rout of riskier assets.

—-

A Chicago Data Center Overheated-and Shut Down Trade in Key Markets Across the Globe

The outage, which lasted for 10 hours, hit CME’s equity, bond and commodity futures. It also offered a warning.

—-

Week Ahead for FX, Bonds: U.S. ISM, ADP Data in Focus as Fed Rate Cut Looks Likely

U.S. ISM surveys on manufacturing and services activity, plus the latest ADP private payrolls, will be watched closely for confirmation that the Federal Reserve could cut interest rates at its next meeting.

—-

Canadian Economy Rebounds by More Than Expected

Canada’s economy recovered far more strongly than anticipated in the latest quarter, pulled out of its decline by a bounce-back in net trade and a surge in defense spending that helped mask weak domestic demand.

—-

The Fed Is Turning the Corner on Profits. It’s Good for the Treasury.

Higher interest rates have brought a tide of red ink to the bank.

  (END) Dow Jones Newswires

  December 01, 2025 01:15 ET (06:15 GMT)
Copyright (c) 2025 Dow Jones & Company, Inc.

Introduction

Diabetes is one of the most common chronic diseases, affecting people of all ages worldwide. According to the latest data for 2021 released by the International Diabetes Federation, there were approximately 537 million cases of diabetes worldwide; 140 million of these were in China, with type 2 diabetes (T2D) accounting for more than 90%, ranking first in the world.¹ Patients with T2D require long-term medication to maintain stable blood glucose control; however, as T2D worsens, insulin production gradually declines, and insulin therapy becomes a major cornerstone of treatment.² Although insulin therapy has health benefits, many patients fail to initiate appropriate intensive insulin therapy in time due to various reasons, including weight gain, the need for education, titration for optimal efficacy, the risk of hypoglycaemia, the necessity of regular glucose monitoring, and the expense of insulin therapy.^3,4

Psychological insulin resistance (PIR) is the term used to describe a patient’s reluctance to initiate insulin therapy.⁵ The negative effects of PIR are complex and multi-dimensional, involving multiple aspects such as psychology, behavior, and clinical outcomes. PIR may cause the best time for patients with T2D to begin insulin therapy to be missed, as well as affect compliance and satisfaction.⁶ Previous studies have shown that even in the presence of diabetes-related complications, 50% of patients who fail to control their blood glucose with oral hypoglycemic drugs only start insulin therapy after a delay of nearly 5 years due to PIR.⁷ Likewise, a study in South Korea showed that due to the impact of RIP, the insulin refusal rate reached 37.5%, and patients in the refusal group had a longer disease duration, more comorbidities, and greater difficulty in maintaining stable blood glucose control.⁸ PIR plays a crucial role in blood glucose control. Poor blood glucose control may reduce patients’ ability to engage in important activities and actions, lower their treatment confidence, and affect their mental health, ultimately leading to a vicious cycle that impacts various aspects of their quality of life.⁹ Therefore, it is necessary to explore the influencing factors of PIR in patients with T2D in order to take targeted measures to reduce the incidence of PIR.

Diabetes stigma (DS) usually refers to the negative emotional experience of DM patients, including labelling, stereotypes, separation, loss of status, and differential treatment.¹⁰ Stigmatisation is a risk factor for PIR, caused primarily by the lack of private injection areas, and may lead to injections being too early or omitted, which may affect treatment compliance.^9,11

Diabetes distress (DD) mostly consists of distress related to lifestyle changes, heightened emotional burden, medical care, and interpersonal communication.¹² Besides increasing the psychological pressure of patients, DD also leads to a decline in the ability of T2D patients to manage their diseases on their own, especially concerns and refusals regarding insulin use, namely PIR, which has an impact on blood glucose regulation.¹³ Furthermore, previous studies displayed that through experiential avoidance,¹⁴ Among patients with T2D, DD has a positive alleviating effect on DS in patients with T2D; that is, a high level of DD is associated with a high level of DS.¹⁵ From another perspective, a study found that DS may also aggravate DD by reducing self-care, self-efficacy and increasing perceived burden.¹⁶ Relevant evidence indicates a close connection between DD and DS. An increase in DD may intensify the perception of DS, and DS would also be magnified over time due to DD.

The belief that one can impact events and subsequently modify conduct is known as self-efficacy (SE).¹⁷ Based on the evidence, SE is an invaluable resource for predicting intention and behaviour related to diabetic self-management, which in turn helps patients adhere to their treatment plans and medication schedules.¹⁸ Research indicates that there may be a relationship between DD and SE, indicating that high SE is a major protective factor against DD and may be able to predict low DD in individuals with T2D.^19,20 SE may be able to alleviate the stigma associated with patients by modulating DD, which in turn lowers PIR.

Social support (SS) is a multifaceted framework that encompasses informational, instrumental, and emotional support.²¹ According to a study of low-income individuals with T2D, low satisfaction with SS was linked to severe DD compared to moderate to high satisfaction.²² A study has shown that high levels of SS can help patients with T2D live an active life, which reduces their PIR.²³

The majority of earlier research on patients with T2D only described the direct relationship between several variables and PIR. Only the direct correlation between various variables and PIR can be evaluated through correlation or regression methods. However, observing indirect effects can provide us with a new perspective on how these different influencing factors interact with each other, better helping prevent PIR. Overall, the directionality and magnitude of some relationships remain uncertain and rarely adjust for socio-economic confounders. To the best of our knowledge, this study is the first to utilise a structural equation model (SEM) to explore the pathways between DS, DD, SE, SS, and PIR in patients with T2D, as well as the direct and indirect effects between variables. The results provide a theoretical basis and intervention strategies to improve PIR in patients with T2D.

Materials and Methods

Study Design and Participants

This study employed a cross-sectional design. In accordance with the STROBE guidelines, convenience sampling was used to select outpatients and inpatients in the Department of Endocrinology of the First Affiliated Hospital of Anhui Medical University (a comprehensive Grade 3A hospital in Hefei, Anhui Province, China) between March and September 2023. The inclusion criteria were age ≥ 18 years old, diagnosed with T2D, willing to give written informed consent, and no mental illness or cognitive impairment. Based on the sampling calculation method of the structural equation model, the study’s sample size should be at least 10 observations per free parameter in the model, or more than 200 cases.²⁴ This research was conducted in accordance with the principles of the Declaration of Helsinki. The Ethics Committee of the First Affiliated Hospital of Anhui Medical University granted consent for this study (approval No. 84230040).

Measurements

Demographic Characteristics

The survey encompassed both general demographic data such as age, sex, residence, educational attainment, marital status, employment status, monthly income, and medical insurance payment, as well as disease-specific data such as duration of diabetes, family history of diabetes, therapeutic method, diabetes-related education, comorbidities, and complications. In this study, the duration of diabetes was measured in years, referring to the period from the first clinical diagnosis of T2D to the date of the survey. The therapeutic method included none, oral hypoglycemic agents (OHA), injection of insulin, or OHA plus injection of insulin.

Insulin Treatment Appraisal Scale (ITAS)

The ITAS was developed by Snoek et al to assess the appraisal of insulin therapy in patients with T2D.²⁵ The scale was fully translated into Chinese by Chen et al, which showed good internal consistency and satisfactory validity in the Chinese population.²⁶ The ITAS consists of two dimensions: Positive Attitude (PA) and Negative Attitude (NA), with a total of 20 items. It is scored using the Likert 5-point method, ranging from 1 (strongly oppose) to 5 (strongly agree). The four positive items scores were reversed when calculating the overall score, ranging from 20 to 100. A higher total score correlated with a more negative evaluation of insulin therapy. The Cronbach’s alpha coefficient of ITAS was 0.86 in this study.

Self-Efficacy for Diabetes Scale (SED)

The SED, created by Lorig et al, aims to assess self-efficacy in diabetic patients.²⁷ It was translated into Chinese by Sun et al, with good reliability and validity.²⁸ 8 items in total were rated on a 10-point response scale from 1 (not at all confident) to 10 (totally confident). The average of the evaluated items determined the final score, ranging from 1 to 10 points, and high scores indicate a high level of self-efficacy. In this study, Cronbach’s α reliability coefficient was found to be 0.98.

Social Support Rating Scale (SSRS)

The SSRS was developed and validated by Xiao, a Chinese researcher, and has been widely used among the Chinese population.²⁹ The scale has been proved to have good reliability and validity in patients with T2D.³⁰ The scale consisted of 10 items composed of three dimensions: objective support (OS), subjective support (SubS), and utilisation of support (US). The total score was the sum of all items and represented the level of social support. The SSRS yielded a total score ranging from 12 to 66 points; scores ≤ 22, 23−44, and 45−66 were classified as low, moderate, and high levels of perceived social support, respectively.³¹ The Cronbach’s alpha for this research scale was 0.86.

Diabetes Distress Scale (DDS)

The DDS was developed by Polonsky et al in 2005 to assess diabetes-related distress.¹² And then Yang and Liu translated it into Chinese and reported Cronbach’s alphas and test-retest reliability.³² The 17-item measure was divided into four categories: emotional burden (EB), interpersonal distress (ID), regimen distress (RD), and physician distress (PD). Distress experienced during the previous month was measured on a 6-point Likert scale, with 1 representing no distress and 6 representing severe distress, ranging from 17 to 102 points. Using the item mean as the cutoff, < 2 were classified as indicating little or no distress, 2−2.9 as moderate distress, and ≥ 3 as high distress.³³ The Cronbach’s alpha for this research scale was 0.91.

Stigma Scale for Chronic Illness (SSCI)

The SSCI was created by Rao et al in 2009 to assess stigma in patients with chronic diseases.³⁴ And the scale was translated into Chinese by Lu et al, showed good internal consistency and convergent validity.³⁵ The scale consisted of 24 items composed of two dimensions: internalized stigma (IS) and enacted stigma (ES). The score ranged from 24 to 120 points, with 1 representing never and 5 representing always. The higher the score, the higher the level of stigma. The Cronbach’s alpha for this research scale was 0.87.

Data Collection

Two skilled investigators collected face-to-face data. Face-to-face data collection was standardized in three steps: (1) pre-collection—participants received standardised instructions, provided informed consent, and were explicitly assured of anonymity and confidentiality; (2) during collection—participants self-completed the questionnaire; clarifications, when requested, were given verbatim from a neutral script. For illiterate or visually impaired individuals, items were read aloud verbatim without prompting, and responses were transcribed exactly; (3) post-collection—each questionnaire was immediately screened for completeness, and any missing data were rectified on site. And the questionnaires were sealed in opaque envelopes. Before the official survey, a pre-survey was conducted to identify potential issues and evaluate the reliability of the scales. Participants were given standardised instructions that helped them overcome their reading challenges, and the goals of the study were explained. Each participant signed an informed consent form and completed the questionnaire anonymously.

Statistical Analysis

All data were independently entered and coded by two researchers using Epidata 3.1. Statistical descriptions, reliability analyses, and correlation analyses were performed utilizing SPSS 26.0. Means and standard deviations were used to convey continuous data, whereas the frequencies and percentages were used to express categorical data. Univariate analysis was conducted using independent sample t-tests and one-way analysis of variance. Pearson’s correlation coefficient was used to show the association between two variables. Significant factors associated with PIR were included in multiple linear regression analysis for further analysis.

PIR, SS, DD, and DS were regarded as latent variables, whereas SE and the linked dimensions of the latent variables were considered observable variables. An SEM was constructed using AMOS 24.0 to determine the total, indirect, and direct effects among the variables. The overall fitness of the model was assessed using the model-fit indices, which includes CMIN/DF, root mean square error of approximation (RMSEA), comparative fit index (CFI), goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), normed fit index (NFI), Tucker-Lewis Index (TLI), and incremental fitting index (IFI). The effects and significance of statistical results among various variables were evaluated using the bootstrap bias-corrected percentile method. A total of 5000 repeated samples were selected to test whether the mediating effect was significant (95% confidence interval does not include zero). Statistical significance was defined as a P value < 0.05.

Results

Basic Participant Characteristics

A total of 306 qualified patients were surveyed, of which 17 questionnaires were excluded due to an inability to complete or invalid answers; the overall response rate was 94.4%. The respondents’ characteristics are presented in Table 1. The average age of the 289 patients was 51.53 years old (standard deviation: 12.61, range: 18–77 years). More than half of the patients were male (n = 178, 61.6%). The majority were urban residents (n = 204, 70.6%), had a family history of diabetes (n = 121, 41.9%), had completed elementary school or less (n = 80, 27.7%), were married (n = 258, 89.3%), were employed (n = 132, 45.7%), had a monthly income of 3000–4999 CNY (n = 104, 36.0%), and had urban residents’ basic health insurance (n = 116, 40.1%). The duration of diabetes was > 10 years (n = 105, 36.3%); most were treated with oral hypoglycaemic drugs combined with insulin injections (n = 150, 51.9%), received diabetes-related education (n = 171, 59.2%), had comorbidities (n = 138, 47.8%), and had complications (n = 74, 25.6%).

Table 1 Baseline Characteristics and Differences in PIR Level (n=289)

Sex, residence, employment status, monthly income, medical insurance payment, duration of diabetes, therapeutic method, diabetes-related education, and comorbidity showed statistically significant differences in PIR levels in patients with T2D (P < 0.05). Compared with participants who had low PIR, those with high PIR were significantly more likely to be female, reside in rural areas, be unemployed, have lower monthly income, pay out-of-pocket, have a shorter diabetes duration, use oral antidiabetic drugs only, lack diabetes-related education, and have no comorbidities. There were no significant differences in the effects of age, educational attainment, marital status, family history of diabetes, or complications (P ≥ 0.05).

Descriptive Statistics for Measured Variables

Table 2 presents the descriptive statistics of the measured variables. The average SE was 7.07 ± 1.56. The average scores of OS, SubS, and US were 9.98 ± 3.08, 24.25 ± 4.08, and 6.64 ± 2.22, respectively. In addition, the average scores of DD and DS were 44.47 ± 17.75 and 35.25 ± 8.64, respectively. The average PIR score was 56.74 ± 11.44. Furthermore, the absolute values of skewness and kurtosis were less than 2 and 4, respectively, which met the conditions of normal distribution.³⁶

Table 2 Descriptive Statistical Results for the Measurement Variables (n=289)

The Relationships Between SE, SS, DD, DS and PIR

The correlation analyses results for SE, SS, DD, DS, and PIR are presented in Table 3. There was a negative association between SE and PIR (r = −0.430, P < 0.01), DD (r = −0.346, P < 0.01), and DS (r = −0.350, P < 0.01). SS showed a negative correlation with DD (r = −0.164, P < 0.01), DS (r = −0.260, P < 0.01), and PIR (r = −0.295, P < 0.01). DD was positively correlated with DS (r = 0.467, P < 0.01) and PIR (r = 0.509, P < 0.01). The DS and PIR scores were significantly associated (r = 0.468, P < 0.01). Multicollinearity was not an issue in this study because the correlation coefficient between the absolute values of the variables ranged from 0.164 to 0.509.

Table 3 The Relationships Between SE, SS, DD, DS and PIR (n=289)

Multiple Linear Regression Analysis

Variables that exhibited significant differences in the univariate and correlation analyses were entered as independent variables, with PIR as the dependent variable, and analyzed using multiple linear regression with a stepwise selection procedure. The final model identified employment status, diabetes-related education, SE, SS, DD, and DS as related to PIR in patients with T2D (P < 0.05). It was determined that the model was statistically significant and the variables included in the model explained 57.7% of the variance (Adjusted R²= 0.577; F = 18.873; P < 0.001). The detailed results are presented in Table 4.

Table 4 Results of Multiple Linear Regression Analysis of PIR (N = 289)

Goodness-of-Fit of the Measurement Model

This study created a preliminary SEM to investigate the relationships between SE, SS, DD, DS, and PIR based on literature reviews, as shown in Figure 1. Path analysis revealed no statistically significant relationship between SS and DS or SE and SS. Given that the model’s CMIN/DF value was 3.100, AGFI value was 0.861, RMSEA value was 0.085, NFI value was 0.899, and TLI value was 0.895, the original model required correction.

Figure 1 Standardized estimates of relationships and effect sizes in the initial model.

To enhance the model’s fit, the following two routes were eliminated: “SS→DS” (P = 0.067), and “SS→SE” (P = 0.161). A modification index of 51.094 was used to modify the model and incorporate the covariances of EB and ID. The modified model is shown in Figure 2. Table 5 presents the results of the goodness-of-fit tests for the initial and modified models.

Table 5 Goodness-of-Fit Test Results Between the Initial Model and the Modified Model (n=289)

Figure 2 Standardized estimates of relationships and effect sizes in the modified model.

Direct and Indirect Effects of the Structural Model

Eight pathways showed statistically significant differences in the path coefficients of the modified model. Table 6 presents the findings for the direct, indirect, and total effects of DS, SE, DD, SS, and PIR. The results showed that SE has a direct effect on PIR (β = −0.321, P < 0.001); DD had the largest positive direct effect on PIR (β = 0.489, P < 0.001) and played a partial mediating role between SE and PIR, with a mediating effect value of −0.190, accounting for 31.1% of the total effect; DS had a positive direct effect on PIR (β = 0.284, P = 0.001) and played a partial mediating role between SE and PIR, with a mediating effect value of −0.052, accounting for 8.5% of the total effect; DS also played a partial mediating role between DD and PIR, and the mediating effect value was 0.124, accounting for 20.2% of the total effect. DD and DS had a chain mediating effect between SE and PIR, with a mediating effect value of −0.048, accounting for 7.9% of the total effect; SS had a direct effect on PIR (β = −0.255, P = 0.007); DD also played a partial mediating role between SS and PIR, with a mediating effect value of −0.105, accounting for 27.2% of the total effect.

Table 6 Standardized Direct, Indirect, and Total Effects in the Modified Model (n=289)

Discussion

This study identified the relevant variables influencing PIR in patients with T2D based on an SEM built using literature reviews. The SEM demonstrated the mechanism of action between these variables. SE and SS had direct and indirect negative effects on PIR in patients with T2D; and DD had both direct and indirect positive effects; DS had only direct positive effects. These findings highlight the significance of SE, DD, and additional elements that may aid in improving PIR in individuals with T2D.

SEM indicated that SE was the second total effect coefficient, except for DD. As demonstrated by numerous earlier research investigations, there was a statistically significant negative association between SE and PIR, which this study reinforced.^23,37–39 According to social cognitive theory, self-efficacy can be used to predict behavior changes related to health, which include goal-setting, mindset, and strategy.⁴⁰ Those with T2D who have high levels of SE tend to view their health changes more positively and have greater adherence to insulin therapy, which helps avoid the development of PIR. As SE is a major factor in lowering PIR, healthcare practitioners should assess and assist patients in enhancing their SE as a preventative measure or educational strategy. A study demonstrated that an eight-week advanced-practice education program for primary-care teams significantly enhanced SE with T2D patients.⁴¹ This suggests that enhancing the training of healthcare professionals to promote SE in T2D patients and alleviating PIR is of great significance.

The results showed that SE can mediate PIR through DD and DS. Therefore, lowering the level of PIR, simultaneously reducing DD and DS, may become an important breakthrough. PIR and DD were positively correlated, indicating that greater PIR corresponded with higher DD levels, in line with prior research.^25,42,43 Research suggests that insulin therapy worsens distress related to emotional burden, such as feeling dazed by injecting oneself, fatigue, and worry about complications, making the disease itself harder to manage.^44,45 Healthcare providers should take proactive steps to check for DD, particularly in the event of complications or changes in therapy.⁴⁶ A recent systematic review found that patients can improve mental health and reduce the distress of diabetes through some psychological interventions, such as cognitive behavioural therapy, guided self-determination, and blood glucose awareness training.⁴⁷

There was a strong positive correlation between DS and PIR. Relationships between DD, SE, and PIR can potentially be mediated by DS. In a study of Turkish teenagers with type 1 diabetes, stigma was found to be a predictor of negative perceptions about insulin treatment.⁴⁸ To adjust their negative perception of insulin, patients should pay attention to their psychological state when managing their diabetes. Additionally, it is critical to provide patients with knowledge about their disease to help them better comprehend their condition, which will lessen stigma.

PIR in individuals with T2D was negatively related to SS, and PIR decreased as SS increased. This correlated with similar research results in the literature.²³ Although SS can come from a variety of sources, information support is the most prevalent, according to a study that examined numerous diabetes phases.⁴⁹ According to a Japanese study, high levels of SS can even buffer negative physiological changes and lower the incidence of diabetic nephropathy.⁵⁰ Our research indicated that SS can also lower DD. A 12-week pilot study noted that SS received during intervention was critical to lowering the stress associated with managing the illness.⁵¹ The pathway analysis of this study showed that SS and PIR were partially mediated by DD. Healthcare professionals considering assessing and identifying as many SS as possible to lower DD levels may be helpful in reducing PIR levels.

Furthermore, advancements in medication and technology may alleviate PIR by reducing the burden of injections and improving adherence. A meta-analysis showed once-weekly basal insulin analogues, such as icodec insulin, significantly reduced the frequency of injection compared with once-daily injections, which may be an important factor in mitigating PIR.⁵² Needle-free insulin administration has been widely used by improving insulin injection devices, reducing pain and skin trauma, and supporting better compliance and satisfaction.⁶ Therefore, future studies can improve the injection frequency and drug delivery device to reduce PIR.

This study had several limitations. First, the capacity to deduce causation was restricted by the cross-sectional study design. While the SEM is useful in demonstrating associations, the relationships identified do not imply causation. Future longitudinal and experimental research must be designed to further investigate the causal relationships between these variables, especially to further verify the relationship between DD and DS. Because it was a cross-sectional study, the reverse pathway (DS→DD) or the interaction with time cannot be ruled out. Second, convenience sampling was used to select the sample, which meant that it was not representative because it came from a hospital. Consequently, it is essential to perform a multicenter survey utilizing a random sampling method in the future. Finally, there could be subjective bias because the data were gathered using self-reported assessments.

Conclusion

This study is the first to investigate the variables associated with PIR in Chinese patients with T2D using the SEM method. SE, SS, DD, and DS were significant determinants of PIR; among these, SE and DD were the most critical variables linked to PIR in individuals with T2D. Elements influencing SE and DD may aid in the creation of intervention plans, assessing SE and DD before intervention, and boosting patient self-assurance in handling their condition, all of which will reduce the likelihood of PIR. Simultaneously, the effects of SS and DS on the PIR should also be considered.

Data Sharing Statement

The data of the study can be obtained by requesting the corresponding author for reasonable reasons.

Ethics Approval and Informed Consent

This study was approved by the Ethics Committee of the First Affiliated Hospital of Anhui Medical University with an approval number of 84230040 and was conducted according to the Helsinki Declaration.

Acknowledgments

We would like to thank all patients with diabetes and hospital staff for their support of this study.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This research was supported by the National Natural Science Foundation of China (82272926); Humanities and Social Sciences Research of Anhui Provincial Higher Education Institutions (SK2020ZD13).

Disclosure

The author(s) report no conflicts of interest in this work.

References

1. Sun H, Saeedi P, Karuranga S, et al. IDF diabetes atlas: global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabet Res Clin Pract. 2022;183:109119. doi:10.1016/j.diabres.2021.109119

2. Ceriello A, deValk HW, Guerci B, et al. The burden of type 2 diabetes in Europe: current and future aspects of insulin treatment from patient and healthcare spending perspectives. Diabet Res Clin Pract. 2020;161:108053. doi:10.1016/j.diabres.2020.108053

3. Daly A, Hovorka R. Technology in the management of type 2 diabetes: present status and future prospects. Diabetes Obes Metab. 2021;23(8):1722–1732. doi:10.1111/dom.14418

4. Davies MJ, Aroda VR, Collins BS, et al. Management of hyperglycaemia in type 2 diabetes, 2022. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetologia. 2022;65(12):1925–1966. doi:10.1007/s00125-022-05787-2

5. Polonsky WH, Fisher L, Guzman S, Villa-Caballero L, Edelman SV. Psychological insulin resistance in patients with type 2 diabetes: the scope of the problem. Diabetes Care. 2005;28(10):2543–2545. doi:10.2337/diacare.28.10.2543

6. Wang W, Men L, Wang Y, et al. Effect of needle-free injection on psychological insulin resistance and insulin dosage in patients with type 2 diabetes. Front Endocrinol. 2024;15:1379830. doi:10.3389/fendo.2024.1379830

7. Rubino A, McQuay LJ, Gough SC, Kvasz M, Tennis P. Delayed initiation of subcutaneous insulin therapy after failure of oral glucose-lowering agents in patients with type 2 diabetes: a population-based analysis in the UK. Diabet Med. 2007;24(12):1412–1418. doi:10.1111/j.1464-5491.2007.02279.x

8. Song Y, Ku BJ, Cho J, Jun Y, Kim B, Nam S. The prevalence of insulin refusal and psychological insulin resistance among Korean patients with type 2 diabetes mellitus. Ann Transl Med. 2019;7(23):760. doi:10.21037/atm.2019.11.77

9. Brod M, Kongsø JH, Lessard S, Christensen TL. Psychological insulin resistance: patient beliefs and implications for diabetes management. Qual Life Res. 2009;18(1):23–32. doi:10.1007/s11136-008-9419-1

10. Link BG, Phelan JC. Conceptualizing stigma. Annu Rev Sociol. 2001;27(1):363–385. doi:10.1146/annurev.soc.27.1.363

11. Jha S, Panda M, Kumar S, et al. Psychological insulin resistance in patients with type 2 diabetes. J Assoc Physicians India. 2015;63(7):33–39.

12. Polonsky WH, Fisher L, Earles J, et al. Assessing psychosocial distress in diabetes: development of the diabetes distress scale. Diabetes Care. 2005;28(3):626–631. doi:10.2337/diacare.28.3.626

13. de Vries L, van der Heijden AAWA, van ‘t Riet E, et al. Peer support to decrease diabetes-related distress in patients with type 2 diabetes mellitus: design of a randomised controlled trial. Bmc Endocr Disord. 2014:14. doi:10.1186/1472-6823-14-21

14. Seo K. The mediating effect of experiential avoidance on the relationship between diabetes distress and self-stigma in people with diabetes mellitus type 2 in Republic of Korea. Healthcare. 2023;11(20):2773. doi:10.3390/healthcare11202773

15. Wang R-H, Lin -C-C, Chen S-Y, Hsu H-C, Huang C-L. The impact of self-stigma, role strain, and diabetes distress on quality of life and glycemic control in women with diabetes: a 6-month prospective study. Biol Res Nurs. 2021;23(4):619–628. doi:10.1177/10998004211009606

16. Yeung NCY, Lee EKP, Kong APS, Leung MKW. “Shame on Me”: exploring the role of self-stigma in psychological outcomes among type 2 diabetes patients in Hong Kong. Int J Behav Med. 2024;31(2):241–251. doi:10.1007/s12529-023-10176-z

17. Bandura A, Adams NE. Analysis of self-efficacy theory of behavioral change. Cognit Ther Res. 1977;1(4):287–310. doi:10.1007/BF01663995

18. Anderson RM, Funnell MM, Fitzgerald JT, Marrero DG. The diabetes empowerment scale: a measure of psychosocial self-efficacy. Diabetes Care. 2000;23(6):739–743. doi:10.2337/diacare.23.6.739

19. Nouwen A, Urquhart Law G, Hussain S, McGovern S, Napier H. Comparison of the role of self-efficacy and illness representations in relation to dietary self-care and diabetes distress in adolescents with type 1 diabetes. Psychol Health. 2009;24(9):1071–1084. doi:10.1080/08870440802254597

20. Lin P-Y, Lee T-Y, Liu C-Y, Lee Y-J. The effect of self-efficacy in self-management on diabetes distress in young people with type 2 diabetes. Healthcare. 2021;9(12). doi:10.3390/healthcare9121736

21. van Dam HA, van der Horst FG, Knoops L, Ryckman RM, Crebolder HFJM, van den Borne BHW. Social support in diabetes: a systematic review of controlled intervention studies. Patient Educ Couns. 2005;59(1):1–12. doi:10.1016/j.pec.2004.11.001

22. Presley CA, Mondesir FL, Juarez LD, et al. Social support and diabetes distress among adults with type 2 diabetes covered by alabama medicaid. Diabet Med. 2021;38(4). doi:10.1111/dme.14503

23. Yu JH, Kim HY, Kim SR, Ko E, Jin HY. Factors influencing psychological insulin resistance in type 2 diabetes patients. Int J Nurs Pract. 2019;25(3). doi:10.1111/ijn.12733

24. Kline RB. Principles and practice of structural equation modelling. Guilford Press;2015.

25. Snoek FJ, Skovlund SE, Pouwer F. Development and validation of the insulin treatment appraisal scale (ITAS) in patients with type 2 diabetes. Health Qual Life Outcomes. 2007;5(1). doi:10.1186/1477-7525-5-69

26. Chen -C-C, Li T-C, Huang C-Y, Chang M-P. Validation of the Chinese version of the insulin treatment appraisal scale. Diabet Res Clin Pract. 2020;170:1. doi:10.1016/j.diabres.2020.108485

27. Lorig K, Stewart A, Ritter P, Gonzalez V, Laurent D, Lynch J. Outcome Measures for Health Education and Other Health Care Interventions. 2455 Teller Road, Thousand Oaks California 91320. United States: Sage Publications; 1996. doi:10.4135/9781452232966

28. Sun SN. Study on Self-management Status and Influencing Factors of Diabetic Patients [Dissertation]. Peking Union Medical College; 2010.

29. Xiao SY. Theoretical basis and research application of ‘social support rating scale’. J Clin Psychiatry. 1994;4(2):98–100.

30. Wang X, Zhang F, Ge Y, Ding Y, Liu T. The associations between social support, self-regulatory fatigue, and health-promoting behaviors among people with type 2 diabetes mellitus: a cross-sectional survey. Front Public Health. 2023;11:1281065. doi:10.3389/fpubh.2023.1281065

31. Zhang Y, Yang N, Bai X, et al. Factors affecting postdialysis fatigue among hemodialysis patients: a multi-group path analysis according to nutritional status. Front Med. 2025;12:1553751. doi:10.3389/fmed.2025.1553751

32. Yang Q, Liu XQ. Reliability and validity of the diabetes distress scale. J Nurs. 2010;17(18):8–10. doi:10.16460/j.issn1008-9969.2010.17.023

33. Fisher L, Hessler DM, Polonsky WH, Mullan J. When is diabetes distress clinically meaningful?: establishing cut points for the diabetes distress scale. Diabetes Care. 2012;35(2):259–264. doi:10.2337/dc11-1572

34. Rao D, Choi SW, Victorson D, et al. Measuring stigma across neurological conditions: the development of the stigma scale for chronic illness (SSCI). Qual Life Res. 2009;18(5):585–595. doi:10.1007/s11136-009-9475-1

35. Lu Q, Deng C, Fu L, et al. Reliability and validity of a Chinese version of the Stigma Scale for Chronic Illness (SSCI) in patients with stroke. Top Stroke Rehabil. 2019;26(4):312–317. doi:10.1080/10749357.2019.1592306

36. MacIniosh R. A program implementing mardia’s multivariate normality test for use in structural equation modeling with latent variables. Educ Psychol Meas. 1997;57(2):357–359. doi:10.1177/0013164497057002013

37. Nam S, Chesla C, Stotts NA, Kroon L, Janson SL. Factors associated with psychological insulin resistance in individuals with type 2 diabetes. Diabetes Care. 2010;33(8):1747–1749. doi:10.2337/dc10-0099

38. Nam S, Nam S, Song Y. Role of self-efficacy in the relationship between patient provider relationships and psychological insulin resistance among patients with type 2 diabetes. J Contemp Diabetes Res. 2014;1(1):1–15.

39. Lim A, Song Y. The role of psychological insulin resistance in diabetes self-care management. Nurs Open. 2020;7(3):887–894. doi:10.1002/nop2.462

40. Bandura A, Locke EA. Negative self-efficacy and goal effects revisited. J Appl Psychol. 2003;88(1):87–99. doi:10.1037/0021-9010.88.1.87

41. Yunis B, Echevarria-Perez P, Morante JJH, Morales-Moreno I. Increasing self-efficacy for the management of patients with type 2 diabetes through an advanced practice education program for primary care professionals. Nurs Rep. 2024;14(4):3830–3846. doi:10.3390/nursrep14040280

42. Polonsky WH, Hajos TR, Dain MP, Snoek FJ. Are patients with type 2 diabetes reluctant to start insulin therapy? An examination of the scope and underpinnings of psychological insulin resistance in a large, international population. Curr Med Res Opin. 2011;27(6):1169–1174. doi:10.1185/03007995.2011.573623

43. Holmes-Truscott E, Skinner TC, Pouwer F, Speight J. Explaining psychological insulin resistance in adults with non-insulin-treated type 2 diabetes: the roles of diabetes distress and current medication concerns. Results from Diabetes MILES—Australia. Prim Care Diabet. 2016;10(1):75–82. doi:10.1016/j.pcd.2015.06.006

44. Fisher L, Polonsky WH, Hessler D. Addressing diabetes distress in clinical care: a practical guide. Diabet Med. 2019;36(7):803–812. doi:10.1111/dme.13967

45. Trief PM, Uschner D, Tung M, et al. Diabetes distress in young adults with youth-onset type 2 diabetes: TODAY2 study results. Diabetes Care. 2022;45(3):529–537. doi:10.2337/dc21-1689

46. Young-Hyman D, de Groot M, Hill-Briggs F, Gonzalez JS, Hood K, Peyrot M. Psychosocial care for people with diabetes: a position statement of the American Diabetes Association. Diabetes Care. 2016;39(12):2126–2140. doi:10.2337/dc16-2053

47. Diribe O, Palmer K, Kennedy A, et al. A systematic literature review of psychological interventions for adults with type 1 diabetes. Diabetes Ther. 2024;15(2):367–380. doi:10.1007/s13300-023-01513-2

48. Arda Sürücü H, Baran Durmaz G, Turan E. Does type 1 diabetic adolescents’ fear of stigmatization predict a negative perception insulin treatment? Clin Nurs Res. 2020;29(4):235–242. doi:10.1177/1054773818815258

49. Da Moura Semedo C, Bath PA, Zhang Z. Social support in a diabetes online community: mixed methods content analysis. JMIR DiabeT. 2023;8:1. doi:10.2196/41320

50. Ninomiya H, Katakami N, Matsuoka T, et al. Association between poor psychosocial conditions and diabetic nephropathy in Japanese type 2 diabetes patients: a cross‐sectional study. J Diabetes Investig. 2018;9(1):162–172. doi:10.1111/jdi.12641

51. Misra R, Shawley-Brzoska S, Khan R, Kirk BO, Wen S, Sambamoorthi U. Addressing diabetes distress in self-management programs results of a randomized feasibility study. J Appalach Health. 2021;3(3):68–85. doi:10.13023/jah.0303.06

52. Karakasis P, Patoulias D, Pamporis K, et al. Efficacy and safety of once-weekly versus once-daily basal insulin analogues in the treatment of type 2 diabetes mellitus: a systematic review and meta-analysis. Diabetes Obes Metab. 2023;25(12):3648–3661. doi:10.1111/dom.15259

Shoppers held back from visiting high streets over Black Friday, data shows, amid fears weak consumer spending will put the brakes on economic growth in 2026.

Visitors to all UK shopping destinations were down 2% on Friday and 7.2% compared with the equivalent days last year, according to the monitoring company MRI Software, with locations near central London offices among the few to experience a lift in visits.

While most Black Friday sales are now online, the picture there was also a mixed bag in the run-up to the big weekend. Sales down heavily on Thursday but up on Tuesday, according to the online retail association IMRG.

“The cost of living squeeze appears to be weighing on overall activity,” said Jenni Matthews from MRI.

The lacklustre data came as the consultancy KPMG highlighted soft consumer spending as one factor likely to hold back the economy over the next 12 months.

Despite the fact that most of the £26bn tax-raising impact of Rachel Reeves’s budget would not be felt until later, KPMG suggested cash-strapped households would continue to be cautious, as unemployment ticks up to 5.2%.

“The outlook for growth in 2026 is subdued, reflecting the impact of a cooling labour market and weak household spending,” said KPMG’s chief economist, Yael Selfin, although she pointed to positive “pockets”, including green energy.

“The medium-term picture could improve further if planning reforms unlock housing delivery and uncertainty reduces for investors,” she added, predicting GDP growth of 1% for 2026 and 1.4% for 2027.

This gloomy outlook chimes with two other reports published on Monday, both underlining the downbeat mood among business leaders.

The Confederation of British Industry’s services sector survey, carried out before the budget, showed the fastest decline in optimism about the general business situation for three years, with companies citing rising costs and uncertainty about future demand.

“Looking ahead, businesses expect little near-term relief, with uncertainty about demand and persistent cost pressures set to constrain future hiring and investment plans,” said the CBI’s Charlotte Dendy.

Separately, the Institute of Directors said its economic confidence index, based on a survey of business leaders, was at a near record low of -73 in the run-up to Reeves’s budget, and improved by a whisker to -72 afterwards.

Anna Leach, the IoD’s chief economist, said: “In the weeks running up to the budget, persistent speculation over tax rises kept confidence subdued. And with our snap poll showing that four in five business leaders (80%) view the budget negatively, it is no surprise that confidence remains close to record lows afterwards.”

skip past newsletter promotion

after newsletter promotion

Separately, hospitality businesses said they would take a big hit from business rates changes, forcing them to rein in investment and hiring. They said measures announced in the budget to protect businesses, as Covid-era support comes to an end, were not enough to offset rises linked to the increase in rateable value of their properties.

Under the complex tax system, for many pubs in particular there will be a big increase next year in their rateable value – a key part of the business rates calculation. This is in contrast to many retailers whose rateable value will fall because of poorer trade on high streets.

In her budget speech, Reeves announced that she was introducing “permanently lower tax rates for over 750,000 retail, hospitality and leisure properties”, paid for with higher rates on the biggest retailers, including big online companies.

However, Paul Crossman, the chair of the Campaign for Pubs and the licensee of three pubs in York, said: “In the vast majority of cases it seems that instead of the promised reduction in our bills [our members] will be expected to pay more, in many cases vastly more, once the existing support finishes next April.”

Alex Reilley, the head of the Loungers chain, said for his business there was a mixed picture because some sites were not categorised as pubs, but he added: “Most [hospitality] businesses will be looking at an increase of some description and for our pub sector it could quite easily be an extinction event.”

The government has said it will provide billions of pounds in “transitional relief” to support those hit by big increases in business rates next year but analysts have dismissed this as merely delaying the pain.

Australian dollar rallies after inflation beats forecasts

AUD/USD wrapped up last week on a firm note, settling at 0.6550 for a 1.45% gain – its highest weekly close in eight weeks. The Australian dollar’s rise was fuelled by several supportive factors.

On Wednesday, Australia’s monthly consumer price index (CPI) for October was hotter than expected. Headline inflation rose by 3.8% year-on-year (YoY) in October, and the trimmed mean measure rose by 3.3%. This resulted in the Australian interest rate market moving from pricing in a modest chance of a Reserve Bank of Australia (RBA) rate cut in 2026 to a modest chance of a rate hike. Additionally, the same day saw the Reserve Bank of New Zealand (RBNZ) deliver a hawkish 25 basis point (bp) rate cut, signalling it was likely done cutting interest rates.

Meanwhile, expectations for next week’s Federal Open Market Committee (FOMC) meeting saw a dovish repricing, with the implied probability of a 25 bp cut on 10 December soaring from around 30% in mid-November to approximately 85% today.

This rapid shift in Federal Reserve (Fed) and RBA rate expectations, alongside buoyant risk appetite, enabled AUD/USD and other risk-sensitive currencies to stage an impressive rally by week’s end.

Key factors for AUD/USD outlook

Looking forward, three key factors will determine whether AUD/USD can continue its ascent.

1. Global risk sentiment: must maintain the positive tone from last week. This premise is currently being tested, with US S&P 500 equity futures trading 0.67% lower at 6813, weighed down by a sell-off in Japanese government bonds and the Nikkei225 /on rising expectations of a Bank of Japan (BoJ) rate hike in December.
2. Upcoming US data releases: including September’s core personal consumption expenditures (PCE) inflation, personal income and spending figures, and November’s Institute for Supply Management (ISM) purchasing managers’ indexes (PMIs), will be important. Particular focus will be on the ISM services PMI, given that services account for approximately 70% of US gross domestic product (GDP), and its employment sub-index offers valuable insights into the state of the US labour market.
3. Q3 Australian GDP release: is scheduled for Wednesday and previewed below.

Q3 GDP

Date: Wednesday, 3 December at 11.30am AEDT

In the second quarter (Q2) of 2025, Australian GDP increased by 0.6%, accelerating from 0.3% in the prior quarter, for an annual rate of 1.8%. It was the Australian economy’s 15thconsecutive quarter of growth.

The number was stronger than expected, driven by a sharp rebound in household consumption, providing evidence the RBA’s rate-cutting cycle is gaining traction.

As we await the final partial components that feed into Wednesday’s GDP print, the preliminary forecast is for a rise of 0.7% quarter-on-quarter (QoQ), lifting the annual growth rate to 2.2% – the fastest pace of annual growth since Q1 2023.

AU GDP chain volume measures chart

Introduction

Parkinson’s disease (PD) is a common neurodegenerative disorder.¹ PD is characterized by progressive degeneration of dopaminergic neurons (DaN) in the substantia nigra (SN), leading to hallmark motor symptoms.² Current treatments for PD are largely limited to replacing dopamine levels and activating dopamine receptors. These therapeutics fail to halt degenerative progression.³ Consequently, there is an urgent need to develop novel therapeutic strategies for PD.

Ferroptosis is a unique mechanism of iron-dependent regulated cell death characterized by altered iron homeostasis, accumulation of lipid peroxides, and imbalanced oxidative defense.⁴ The protein transferrin receptor 1 (TfR1), encoded by transferrin receptor (TFRC), and Solute Carrier Family 11 member 2 (SLC11A2), plays a key role in cellular iron import and influences systemic iron homeostasis.⁵ Iron accumulation in the brain worsens with age and is considered a hallmark of PD.⁶ It is documented that neurons acquire most of their iron via the transferrin-TfR1 system.⁷ Neurons regulate iron levels by minimal sequestration or efficient export via Ferroportin1 (FPN).⁸ Although TfR1 is broadly involved in iron uptake, its upregulation has been reported in various ferroptosis models, suggesting it may contribute to ferroptosis susceptibility by promoting intracellular iron accumulation.⁹ Acyl-CoA synthetase long-chain family member 4 (ACSL4), an enzyme esterifying CoA into specific polyunsaturated fatty acids, may trigger phospholipid peroxidation and induce ferroptosis.¹⁰ Inhibition of ACSL4 may reduce lipid peroxides and ameliorate PD.¹¹ Glutathione peroxidase 4 (GPX4) is a key suppressor of ferroptosis.¹² Abnormal iron accumulation in the SN and elevated oxidative stress are key pathological features of PD.^8,13 In a 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) induced mouse model, midbrain exhibits dysregulation iron metabolism, increased ACSL4 expression and impaired GPX4 activity.¹⁴ These findings highlight the contribution of ferroptosis to PD progression and the therapeutic potential of targeting ferroptosis-related pathways.

The p53 protein, encoded by the tumor protein p53 (TP53, or p53) gene, has been appeared to be a highly attractive target for anti-cancer research.¹⁵ TP53 is recognized as a tumor suppressor and executes its diverse functions, including cellular cycle arrest, DNA damage repair, metabolic homeostasis, cell senescence, autophagy, and apoptosis.¹⁶ Various researches have implicated recently that p53 has an association with ferroptosis.¹⁷ Cardiomyocytes challenged by doxorubicin exhibits iron overload and lipid peroxide accumulation in a p53-dependent manner.¹⁸ The activation of p53 results in iron homeostasis dysregulation through increasing TfR1 expression.¹⁸ Importantly, emerging studies have shown that p53-mediated ferroptosis may be involved in neuronal death in PD models, suggesting that p53 is a potential contributor to neurodegeneration via ferroptosis pathways.^19,20 Targeting the p53-ferroptosis signaling shows promise for PD treatment.

Emodin (EMD), a natural anthraquinone compound, is derived from multiple Chinese medicinal herbs including Cassia obtusifolia, Aloe vera, Polygonum cuspidatum, Polygonum multiflorum, and Rheum palmatum.²¹ Emerging evidences show that EMD confers various beneficial actions, including anti-bacterial, anti-inflammatory, anti-ferroptosis and neuroprotective effects.²² It is reported that EMD has a beneficial effect on various neurodegenerative diseases, including PD and Alzheimer’s disease.²² Although recent researches suggest EMD has positive effects on neuroprotection via a wide range of biological targets, its role in controlling neurodegenerative diseases and its pharmacological mechanism of action require further elucidation.

Here, we hypothesized that p53 inhibition-mediated regulation of ferroptosis signaling is the primary pharmacological mechanism of EMD. We explored whether EMD protects against dopaminergic neurodegeneration and clarified whether the beneficial effects of EMD are associated with p53-ferroptosis signaling inhibition.

Materials and Methods

Materials and Reagents

Stock solutions of 10 mM EMD (Selleck Chemicals, S2295, Purity: 99.75%), 400 mM 1-methyl-4-phenylpyridinium (MPP⁺, Aladdin, N137206), 10 mM RSL3 (Med Chem Express, HY-100218A), 1 mM Erastin (Med Chem Express, HY-15763) in DMSO, and 50 mM C16-Ceramide (C16, MCE, HY-100354) in ethanol were used. We prepared small aliquots (30 μL) of each drug and stored them at −20°C. For the cell experiments, the stock solutions were diluted to a specific concentration of 2 mM for MPP⁺, 3 μM for RSL3, 5 μM for Erastin, and 50 μM for C16.

The Cell Counting Kit-8 (CCK-8), Lipid Peroxidation malondialdehyde (MDA) Assay Kit, and MPTP were obtained from Beyotime Biotechnology (Shanghai, China). The Cell Total Iron Colorimetric Assay Kit was obtained from Elabscience (Wuhan, China). MitoSOX Red was purchased from MedChem Express (Shanghai, China). FerroOrange was purchased from Dojindo (Kumamoto, Japan). Antibodies against anti-TfR1, anti-ACSL4, anti-GPX4, anti-FPN, anti-FTH1, anti-β-Actin, anti-TH, anti-LPCAT3, and Alexa Fluor 594-labeled secondary antibody were purchased from Abclonal Technology Co., Ltd. (Wuhan, China). The anti-p53 antibody was obtained from Cell Signaling Technology (Danvers, MA, USA).

Obtaining Potential Targets of EMD and PD

The Canonical SMILES of EMD, available on PubChem (https://pubchem.ncbi.nlm.nih.gov/), was imported into the similarity ensemble approach (SEA, https://sea.bkslab.org/) and Swiss Target Prediction (Swiss, http://www.swisstargetprediction.ch/) to predict the EMD-related targets. Potential targets of Parkinson’s disease (PD) were identified using online Mendelian inheritance in Man (OMIM, https://www.omim.org/) and the GeneCards database (http://www.genecards.org/). We focused on the intersection between the two gene sets for subsequent analysis.

Network Construction and Enrichment Analysis

Overlapping genes for EMD and PD targets were introduced into the STRING database (https://cn.string-db.org/) to construct protein-protein interactions (PPI). The Cytoscape software (version 3.11.0) was used to create a visual network.

For Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, gene lists containing 91 shared targets were entered into a Database for Annotation, Visualization, and Integrated Discovery (DAVID, https://david.ncifcrf.gov/). The results were visualized using R package.

Molecular Docking

Autodock4 software was employed to ascertain the binding possibility of EMD with the core proteins p53 (PDB ID: 8DC6) and TfR1 (PDB ID: 6Y76) for docking simulation. PyMOL 2.6.0 was employed to adjust the protein molecules, including removing water molecules and extraneous ligands and adding hydrogen atoms. AutoDock Tools 1.5.6, was used to simulate molecular docking. Finally, the results were visualized using PyMOL version 2.6.0.

Molecular Dynamic (MD) Simulation

YASARA Structure software was employed to perform molecular dynamics simulation in 80 ns running time. The Root-Mean-Square Deviation (RMSD) and Root-Mean-Square Fluctuation (RMSF) results were used for assessing structural stability.

Cell Culture

The human SH-SY5Y neuroblastoma cell line was procured from Procell Life Science & Technology Co., Ltd. (Wuhan, China). The cells were cultivated in DMEM/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (Gibco, USA) and 1% streptomycin-penicillin solution (Beyotime, China) and maintained at 37°C in a 5% CO₂ incubator. Cells were seeded at a density of 5×10⁵ cells/well in 6-well plates pre-treated with poly-D-lysine (Beyotime, China), followed by co-culturing with EMD (2 μM, 5 μM, and 10 μM), MPP⁺ (2 mM), and C16 (50 μM) for 12 h in the RT-qPCR assay and for 24 h in other assays.

Cell Viability Assay

The CCK-8 assay was conducted to analyze the viability of SH-SY5Y cells treated with various concentrations of EMD (1–80 μM). The cells were incubated at 37°C for 1 h, followed by treatment with 10% CCK-8. The absorbance was measured at 450 nm using a SpectraMax i3x microplate reader (Molecular Devices, CA, USA).

Animal Modeling and Treatment

All animal procedures used in this study were conducted in compliance with the guidelines of the Animal Experimental Ethics Committee of Shanghai Medical School, Fudan University, China (Approval Number: 202411013Z). C57BL/6J mice (10 weeks old, male, 24–26 g) were purchased from Shanghai Jihui Laboratory Animal Care Co., Ltd. (Shanghai, China) and kept under the specific pathogen-free conditions that controlled a 12h light/dark cycle at a temperature of 22–23°C. To generate an acute MPTP model, mice were randomly divided into four groups according to body weight: normal control group (CON, n = 12), MPTP model group (MPTP, n = 12), low-dose EMD treatment group (MPTP + EMDL, 50 mg.kg⁻¹ i.g. EMD, n = 12), and high-dose EMD treatment group (MPTP + EMDH, 100 mg.kg⁻¹ i.g. EMD, n = 12). The EMD treatment groups were administered the corresponding doses once daily for eight days. Subsequently, the mice were intraperitoneally administered MPTP at a dose of 20 mg.kg⁻¹ body weight on the sixth day after EMD administration. The CON group received an equivalent volume of normal saline. After the behavioral tests, the mice were euthanized.

Behavioral Tests

To evaluate spontaneous locomotor activity and coordination ability of the mice, open-field and rotarod tests were conducted. The mice were trained for adaptive exercise for two days prior to the test. For the open field test, a SuperFlex open field system (40 × 40×40 cm; Omnitech Electronics Inc. Columbus, OH) was employed for 3 min to monitor mouse activity, and fusion system software (Omnitech Electronics, Inc.) was used to capture the animals’ real-time position and quantify locomotor parameters. For the rotarod test, mice were placed on the rotarod and the speed was set from 20 to 30 rpm for 1 min. The time at which the mice fell from the rod and the total distance were recorded.

Western Blotting

Whole-protein extracts from SH-SY5Y cells and SN of mice were initially separated by 10% SDS-PAGE and then transferred to PVDF membranes equilibrated with ethanol. The membranes were blocked with 5% bovine albumin for 1 h and subsequently probed with primary antibodies against anti-p53 (1:1000 dilution), anti-TfR1 (1:1000 dilution), anti-ACSL4 (1:1000 dilution), anti-GPX4 (1:1000 dilution), anti-TH (1:1000 dilution), anti-LPCAT3 (1:1000 dilution), anti-FPN (1:1000 dilution), anti-FTH1 (1:1000 dilution), anti-β-Actin (1:50,000 dilution) at 4°C overnight, followed by HRP-labeled secondary antibodies (1:5000 dilution) for 1 h at room temperature. The protein bands were detected using Bio-Rad Image Lab Touch software and analyzed using ImageJ software. The final protein expression levels were normalized to that of β-Actin.

RT-qPCR Assay

Total RNA from mouse SN and SH-SY5Y cells was extracted using a SteadyPure Universal RNA Extraction Kit (Accurate Biology, Hunan, China). First-strand cDNA was synthesized and subjected to quantitative real-time PCR (qRT-PCR) analysis using SYBR Green Pro Taq HS Mix (Accurate Biology, Hunan, China). Data were analyzed on a QuantStudio 6 Flex system (Applied Biosystems, USA) using the 2^−ΔΔCt method. β-Actin was used to normalize gene expression values. The primer sequences used are listed in Table 1.

Table 1 The Sequences of Primer in RT-qPCR Assay

Immunohistochemistry (IHC)

For IHC staining, the mice brain paraffin sections were prepared and incubated with anti-TH (1:200 dilution) overnight at 4°C, followed by incubation with an HRP-conjugated secondary antibody. DAB reagent was used for visualization. The sections were then counterstained with hematoxylin. Finally, images were photographed using a scanning microtome (KF-FL-005, KFBIO, China).

Midbrain MDA and Iron Content Determination

Midbrain samples were homogenized by sonication, followed by centrifugation at 12,000 × g for 10 min. The supernatants were collected for subsequent analyses. The lipid Peroxidation MDA Assay Kit and Cell Total Iron Colorimetric Assay Kit were used to assay midbrain MDA and iron content, following the manufacturer’s instructions.

Intracellular Superoxide Assay

Superoxide levels were measured using MitoSOX Red (MedChem Express, Shanghai, China), following the manufacturer’s instructions. Briefly, SH-SY5Y cells were washed with PBS and stained with 5 μM MitoSOX Red reagent for 30 min in an incubator at 37°C. The cells were then rinsed with PBS for 3 times. Images were captured at a wavelength of 594 nm using an ImageXpress Pico scanner (Molecular Devices, USA).

Immunofluorescence Staining

For Immunofluorescence analysis, SH-SY5Y cells were fixed in 10% formalin and prepared to incubate with a primary antibody overnight at 4°C, followed by incubation with Alexa Fluor 594-labeled secondary antibodies for 1 h at room temperature. The cells were counterstained with DAPI (2 μg.mL⁻¹, Yeasen, Shanghai, China). Images were acquired using an ImageXpress Pico scanner (Molecular Devices).

Flow Cytometry

SH-SY5Y cells were collected after treatment with MPP⁺ and EMD followed by digestion in a single-cell suspension. The harvested cells were treated with 1 μM FerroOrange for 30 min in an incubator at 37°C. Flow cytometric analysis was performed using an Attune NxT Flow Cytometer (Life Technologies). The data were analyzed using Flowjo v10.8.1 software.

Result Presentation

All data are presented as the mean ± SEM. Due to the exploratory nature of this study and the limited sample size (n=3–4), no formal statistical comparisons were performed. The results are intended to describe observed trends and generate hypotheses for future validation.

Results

Network Pharmacology Identifies the Targets of EMD in the Treatment of PD

The chemical structure of EMD is shown in Figure 1A. In total, 11,791 PD-related genes were identified from the OMIM and GeneCards databases, and 117 potential targets of EMD were predicted using the Swiss and SEA databases. Ninety-one therapeutic targets were identified by intersecting these two gene sets (Figure 1B). The PPI network showed that TP53 had the highest number of connections (114), suggesting that TP53 is likely the key target of EMD (Figure 1B). GO enrichment analysis highlighted pathways such as iron ion binding, NADP⁺ activity, mitochondrial function, and reactive oxygen species (ROS) metabolism (Figure 1C). Similarly, KEGG enrichment analysis identified pathways including the p53 signaling pathway, Alzheimer’s disease, and pathways of neurodegeneration involving multiple diseases (Figure 1D). These findings suggest that EMD may exert anti-oxidative stress and anti-ferroptosis effects, highlighting its potential as a therapeutic agent for neurodegenerative diseases. To verify the above results, SH-SY5Y cells were treated with EMD for 24 h and the expression of ferroptosis-related genes and proteins was collected. The results showed that EMD reduced the expression of TP53, TFRC, SLC11A2, and ACSL4, while upregulating the expression of FPN, GPX4, and FTH1 (Figure 1E). Similarly, EMD reduced the protein expression of p53, TfR1, and ACSL4 while increasing GPX4 levels (Figure 1F and G). However, FPN expression did not increase significantly after EMD treatment, but FTH1 expression showed a slight increase (Figure 1F and G). These results suggest that EMD exerts anti-ferroptosis effects and alleviates dopaminergic neurodegeneration.

Figure 1 Network pharmacology explores the targets of EMD in the treatment of PD. (A) The structure of EMD. (B) The PPI network of the overlapped genes between EMD and PD. (C) KEGG enrichment analysis of the overlapped genes. (D) GO enrichment analysis of the overlapped genes. (E) Relative mRNA levels of ferroptosis-related genes in SH-SY5Y cells (n=3). (F and G) Western blot assay validated that EMD effectively depressed the expression of ferroptosis-related protein in SH-SY5Y cells (n=3).

EMD Ameliorates MPTP-Induced Dyskinesia and Dopaminergic Neuronal Damage in Mice

To explore the neuroprotective effects of EMD, we evaluated the behavioral and neural status of MPTP-induced mice following EMD treatment. The in vivo experimental design timeline is shown in Figure 2A. Body weight decreased after MPTP intervention, whereas the high-dose EMD group showed alleviated weight loss (Figure 2B). Motor behavior was evaluated using the open field and rotarod tests. MPTP treatment significantly induces hypokinesia. The total distance decreased in both the open field test and rotarod test, and the changes were rescued by high-dose EMD administration (Figure 2C–E). TH, the rate-limiting enzyme for dopamine synthesis, is the hallmark of DaN.²³ Mice subjected to MPTP administration displayed a decrease in the number of DaN in the SN and the level of the TH protein. EMD treatment mitigated these changes in a dose-dependent manner (Figure 2F–H). Collectively, these findings indicate that EMD exerts neuroprotective effects against MPTP-induced neural damage.

Figure 2 EMD ameliorates dyskinesia and dopaminergic neuronal damage in mice. (A) Schedule and timeline for the experimental design in vivo. (B) The change of body weight during the animal experiment. (C and D) Representative movement images and total distance in the open field test. (E) Total distance in the rotarod test. (F) Immunohistochemistry for TH in the SN of mice. Scale bars: 200 μm. (G and H) Western blot assay validated that EMD effectively protected against MPTP-induced DaN damage.

EMD Rescues the Dysregulation of p53 and Curbs the Ferroptosis in vivo

Ferroptosis has been regarded as one of the critical factors contributing to neurodegeneration in PD, and the inhibition of ferroptosis may have therapeutic potential.²⁴ While measuring the expression of ferroptosis-related genes, we observed that MPTP increased the levels of TFRC, ACSL4 and LPCAT3 in the SN, but did not affect the gene expression levels of SLC11A2, GPX4, FTH1 and SLC7A11. However, the high-dose EMD group exhibited lower TFRC, ACSL4, and LPCAT3, and higher GPX4 and FTH1 mRNA expression (Figure 3A–G). Additionally, MPTP treatment induced higher midbrain MDA and ferric ion levels than in the CON group (Figure 3H and I). This finding suggests that an imbalance between iron homeostasis and lipid peroxidation occurs in the SN. Conversely, EMD treatment reduced the changes in midbrain MDA and iron contents (Figure 3H and I). Similarly, ferroptosis-related protein analysis showed increases in p53, TfR1, and ACSL4 levels along with decreased GPX4 expression. However, EMD treatment mitigated these alterations to varying extents (Figure 3J–O). These results indicate that EMD counteracted ferroptosis in MPTP-treated mice.

Figure 3 EMD rescues the dysregulation of p53 and curbs the ferroptosis in vivo. (A–G) Relative mRNA levels of ferroptosis-related genes in MPTP-induced mice. (H) The assay of MDA content in MPTP-induced mice midbrain. (I) The assay of iron content in MPTP-induced mice midbrain. (J-O) Western blot assay validated that EMD effectively rescued the dysregulation of p53 and inhibited the ferroptosis in mice midbrain.

EMD Attenuates Ferroptosis Induced by MPP+ in SH-SY5Y Cells

Subsequently, we assessed whether EMD attenuated ferroptosis in MPP⁺-induced SH-SY5Y cells. According to CCK8 results, EMD (10 μM) did not affect cell viability (Figure 4A). Both RSL3 and Erastin, ferroptosis inducers that impair cell survival, were relieved by EMD intervention (Figure 4B and C). Furthermore, EMD suppressed the overaccumulation of ROS in SH-SY5Y cells subjected to MPP⁺ insult (Figure 4D and E). These findings suggest that EMD can ameliorate oxidative damage in cells. Meanwhile, the TfR1 immunofluorescence assay showed that EMD reduced TfR1 expression compared with that in the MPP⁺ group (Figure 4D and E). Additionally, flow cytometry analysis of FerroOrange showed that EMD reversed MPP⁺-induced ferrous iron accumulation (Figure 4F). These results indicate that EMD treatment prevents MPP⁺ induced ferrous iron accumulation and ferroptosis in vitro.

Figure 4 EMD attenuates ferroptosis induced by MPP+ in SH-SY5Y cells. (A) Cell viability of SH-SY5Y under different concentrations of EMD stimulation. (B and C) Cell survival of SH-SY5Y under ferroptosis inducers insult. (D and E) Fluorescence staining for MitoSOX or TfR1 in SH-SY5Y cells. Scale bar: 100 μm. (F) Flow cytometry analysis revealed that EMD rescued the imbalance of iron homeostasis in MPP+-induced SH-SY5Y cells.

EMD Rescues the Dysregulation of p53 and Curbs the Ferroptosis in vitro

Given the observed effects of EMD on the regulation of cell iron content and ferroptosis, we investigated the mechanisms of EMD in vitro. In line with the findings obtained from the mouse SN, we confirmed that MPP⁺ had an impact on the expression of TFRC and ACSL4 (Figure 5A and B). EMD treatment relieved these alterations in gene expression to varying degrees (Figure 5A and B). EMD treatment relieved the exacerbated gene expression of TP53 induced by MPP⁺ exposure (Figure 5C). These findings were further validated by detection of cellular ferroptosis-related protein expression (Figure 5D–I). MPP⁺ exposure increased the protein expression of p53, TfR1, ACSL4, and LPCAT3, and decreased GPX4 expression, which was prevented by EMD treatment (Figure 5D–I). Collectively, these results suggest that EMD counteracts ferroptosis induced by MPP⁺.

Figure 5 EMD rescues the dysregulation of p53 and curbs the ferroptosis in vitro. (A and B) Relative mRNA levels of ferroptosis-related genes in SH-SY5Y cells under MPP+ insult. (C) Relative mRNA levels of TP53 in SH-SY5Y cells under MPP+ insult. (D–I) Western blot assay validated that EMD effectively rescued the dysregulation of p53 and ferroptosis-related protein expression.

The Activation of p53 Curbs EMD Inhibiting MPP+‑Triggered Ferroptosis in SH-SY5Y Cells

The binding energy of the small molecule EMD and the proteins (p53 and TfR1) were calculated by Autodock4, resulting in values of −3.28 and −5.15 kcal/mol, respectively. Molecular docking revealed that EMD possibly interacted with p53 and TfR1 (Figure 6A). The RMSD remained within 2–4 Å, suggesting a stable complex structure throughout the simulation (Figure 6B). Most residues showed minimal fluctuation (Figure 6C). Subsequently, we verified whether EMD-induced p53 downregulation mediated ferroptosis inhibition under MPP⁺/MPTP-induced conditions. C16, a selective p53 activator, was used to activate p53 signaling in vitro. EMD reduced TFRC and ACSL4 gene expression in SH-SY5Y cells exposed to MPP⁺. Nevertheless, this suppressive action was attenuated when the cells were co-cultured with C16 (Figure 6D and E). Furthermore, C16 exposure exacerbated the dysregulation of p53, TfR1, and ACSL4 proteins in SH-SY5Y cells induced by MPP⁺ (Figure 6F–I), but it almost invalidated the inhibitory effect of EMD on MPP⁺-mediated upregulation of p53, TfR1, and ACSL4 protein expression (Figure 6F–I). These findings suggest that EMD protect against MPP⁺-induced ferroptosis, possibly through p53 inhibition.

Figure 6 The activation of p53 curbs EMD inhibiting MPP+‑triggered ferroptosis in SH-SY5Y cells. (A) Molecular docking simulations for the binding possibility of EMD and p53 or TfR1. (B) RMSD plot showing the backbone stability of the protein-ligand complex over 80 ns. (C) RMSF plot indicating the residue-wise flexibility. (D and E) Relative mRNA levels of TFRC, and ACSL4 in SH-SY5Y cells under MPP+ or C16 exposure. (F–I) Western blot assay validated that p53 activation attenuated the suppressive effect of EMD on ferroptosis.

Discussion

The role of EMD in controlling dopaminergic neurodegeneration and its underlying mechanisms remains unclear. In this study, EMD alleviated MPTP-triggered ferroptosis, DaN cell death, and neurodegeneration in mice. Mechanistically, EMD inhibited p53 expression and suppressed the ferroptosis pathway in the midbrain of the mice. Our findings suggest that EMD may exert beneficial effects on controlling dopaminergic neurodegeneration.

EMD has been reported to exert an anti-ferroptosis effect and protects against neurodegenerative insults in mice.^22,25 Our network pharmacology analysis indicated that EMD shares 91 targets with potential PD genes, with TP53 being one of the most critical. These targets were investigated using GO and KEGG enrichment analyses. These findings indicate that EMD is associated with p53 signaling, iron-binding, and neurodegenerative diseases. Moreover, molecular docking analysis showed that EMD possibly binds to p53 and TfR1 proteins. Molecular Dynamics Simulation further confirms that the interactions between EMD and p53 remained stable over time. In animal studies, EMD treatment diminished body weight loss in the MPTP group. A slight weight loss after MPTP injection may occur due to decreased food and water intake.²⁶ Treatment of high dose EMD relieved hypokinesia and neurodegeneration in mice, as evidenced by elevated total distance in open field test and rotarod test, improved SN pathologic phenotypes, and enhanced expression of TH. These findings indicate that EMD have potential therapeutic effects against neurodegenerative diseases.

It is well-established that ferroptosis in midbrain is a key factor in PD pathogenesis.²⁷ Inhibiting ferroptosis signaling in brain holds promise as a therapeutic strategy for PD and other brain diseases.¹³ TfR1, responsible for importing iron into cells, is thought to enhance ferroptosis susceptibility by facilitating intracellular iron accumulation.⁹ ACSL4, producing PUFAs to trigger ferroptosis, is a critical gene in the pathogenesis of PD.¹¹ We found that EMD treatment alleviated midbrain lipid peroxidation and accumulation of iron. EMD treatment decreased the expression of TFRC, ACSL4, and LPCAT3 and increased GPX4 expression in the mouse SN. Consistently, EMD treatment decreased the protein expression of TfR1 and ACSL4 while increasing GPX4 expression in the SN of MPTP-induced mice, highlighting its protective effects against ferroptosis in dopaminergic neurodegeneration.

The p53-mediated cell death contributes to dopamine neuron loss and impairs cell survival.²⁸ The p53 signaling pathway regulates iron metabolism, and its imbalance disrupts cellular iron homeostasis.¹⁸ It is reported that Ubiquitin-specific protease 7 upregulates TfR1, inducing ferroptosis via p53 activation.²⁹ In a brain stroke rat model, the brain undergoes ferroptosis due to increased p53 expression and decreased GPX4 activity.³⁰ p53 may modulate ferroptosis responses in the presence of ferroptosis inducers or high concentration ROS, suggesting that the regulation of p53 is potential to control ferroptosis.¹⁷ Additionally, EMD has been reported to attenuate brain damage by suppressing p53.³¹ Consistently, our findings show that EMD acts as both a p53 inhibitor and a ferroptosis modulator. Thus, we hypothesized that EMD regulates p53-ferroptosis signaling and defends against dopaminergic neurodegeneration.

In vitro, EMD suppressed the expression of TP53, TFRC, and ACSL4, while increasing the expression of FPN, GPX4, and FTH1 in SH-SY5Y cells. Similarly, EMD repressed the protein expression of p53, TfR1, and ACSL4 while increasing GPX4 levels. However, FPN expression did not increase significantly after EMD treatment, but FTH1 expression showed a slight increase, suggesting a shift toward intracellular iron sequestration rather than export. Similarly, these results were equally validated when the cells were subjected to MPP⁺ insult. The alterations in gene and protein expression were almost reversed by EMD under MPP⁺ conditions. These findings indicated that EMD suppressed p53 expression and regulated ferroptosis. EMD treatment improved cell survival in response to the ferroptosis inducers RSL3 and Erastin, suggesting its therapeutic potential in ferroptosis regulation. Furthermore, EMD treatment markedly inhibited cellular superoxide following MPP⁺ stimulation. However, MPP⁺ elevates intracellular iron, but EMD does not significantly reduce these levels. Although total cellular iron levels were not significantly reduced in vitro, the changes in ferroptosis-associated proteins and the in vivo iron reduction together support the conclusion that EMD modulates ferroptosis at least in part via iron metabolism regulation. In addition, the observation that MPP⁺ simultaneously increases intracellular iron levels and TfR1 expression suggests a pathological disruption of iron regulation mechanisms. This paradoxical upregulation of TfR1 may result from MPP⁺-induced oxidative stress or p53 involvement. Moreover, the EMD-induced inhibition of TFRC and ACSL4 expression was counteracted by C16. Similarly, C16 counteracted EMD-induced suppression of p53, TfR1, and ACSL4 protein expression, suggesting that EMD regulates ferroptosis signaling, possibly through p53 inhibition.

Our study has some limitations. Firstly, the small sample size (n<5) precludes formal statistical analysis and limits the generalizability of the findings. The results should therefore be interpreted as preliminary observations. Future studies with expanded sample size are needed to confirm these trends and explore their mechanistic basis. Secondly, the 80 ns simulation timeframe provided initial insights into ligand binding stability, longer simulations (≥100 ns) would further consolidate these observations. Future studies will prioritize extended simulations using high-performance computing resources. Thirdly, our study focused on the therapeutic effects of EMD in acute MPTP-induced mice. However, the preventive effects of EMD in acute and chronic PD models require further investigation. Additionally, our research has shown that EMD curbs ferroptosis signaling by modulating the perturbation of ferroptosis molecular expression through the inhibition of p53. Although TfR1 is recognized as an upstream regulator and a hallmark of ferroptosis, further research is required to determine whether the effect of EMD on ferroptosis depends on the p53-TfR1 axis. Furthermore, we propose that the neuroprotective effects of EMD in vivo may result from its regulation of the p53-ferroptosis pathway in DaN cells. This result implicates the possible occurrence of analogous EMD events in other cells, including microglia, astrocytes, and oligodendrocytes, which probably mediate the anti-neurodegenerative effects of EMD.

Conclusions

In conclusion, our exploratory study provides preliminary evidence that EMD may be a natural regulator of the p53-ferroptosis pathway. This compound may alleviate acute MPTP-induced Parkinsonism. The potential mechanism of action of EMD may involve suppression of ferroptosis signaling in DaN via p53 inhibition. These findings suggest the potential use of EMD as a therapeutic agent in PD management.

Abbreviations

ACSL4, Acyl-CoA synthetase long-chain family member 4; C16, C16-Ceramide; CCK-8, Cell Counting Kit-8; DaN, dopaminergic neurons; DAVID, the Database for Annotation, Visualization, and Integrated Discovery; EMD, Emodin; FPN, Ferroportin1; GO, Gene Ontology; GPX4, glutathione peroxidase 4; KEGG, Kyoto Encyclopedia of Genes and Genomes; MPP⁺, 1-methyl-4-phenylpyridinium; MPTP, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; OMIM, online Mendelian inheritance in man; PD, Parkinson’s disease; PPI, protein-protein interaction; qRT-PCR, quantitative real-time polymerase chain reaction; RMSD, Root-Mean-Square Deviation; RMSF, Root-Mean-Square Fluctuation; ROS, reactive oxygen species; SEA, Similarity ensemble approach; SN, substantia nigra; Swiss, Swiss Target Prediction; TfR1, transferrin receptor 1 protein.

Acknowledgments

We thank Dr. Ling Aye for his review of the manuscript. We thank the staff of Dr. Jingcheng Dong (Huashan Hospital, Fudan University) for assistance with this study.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This research was supported by the National Natural Science Foundation of China [grant number: 82073072]. During the revision process, we performed additional Western blot experiments to strengthen our findings, which required the purchase of specific antibodies. These experimental costs were supported by: the National Natural Science Foundation of China [grant number: 82471270].

Disclosure

The authors report no conflicts of interest in this work. This paper is available as a preprint on SSRN at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5073906.

References

1. Morris HR, Spillantini MG, Sue CM, Williams-Gray CH. The pathogenesis of Parkinson’s disease. Lancet. 2024;403(10423):293–304. doi:10.1016/S0140-6736(23)01478-2

2. Ye H, Robak LA, Yu M, Cykowski M, Shulman JM. Genetics and pathogenesis of Parkinson’s syndrome. Annu Rev Pathol. 2023;18:95–121. doi:10.1146/annurev-pathmechdis-031521-034145

3. Weintraub D, Aarsland D, Chaudhuri KR, et al. The neuropsychiatry of Parkinson’s disease: advances and challenges. Lancet Neurol. 2022;21(1):89–102. doi:10.1016/S1474-4422(21)00330-6

4. Sun S, Shen J, Jiang J, Wang F, Min J. Targeting ferroptosis opens new avenues for the development of novel therapeutics. Signal Transduct Target Ther. 2023;8(1):372. doi:10.1038/s41392-023-01606-1

5. Xiao X, Moschetta GA, Xu Y, et al. Regulation of iron homeostasis by hepatocyte TfR1 requires HFE and contributes to hepcidin suppression in β-thalassemia. Blood. 2023;141(4):422–432. doi:10.1182/blood.2022017811

6. Lu LN, Qian ZM, Wu KC, Yung WH, Ke Y. Expression of iron transporters and pathological hallmarks of Parkinson’s and Alzheimer’s diseases in the brain of young, adult, and aged rats. Mol Neurobiol. 2017;54(7):5213–5224. doi:10.1007/s12035-016-0067-0

7. Ward RJ, Zucca FA, Duyn JH, Crichton RR, Zecca L. The role of iron in brain ageing and neurodegenerative disorders. Lancet Neurol. 2014;13(10):1045–1060. doi:10.1016/S1474-4422(14)70117-6

8. Yao Z, Jiao Q, Du X, et al. Ferroptosis in Parkinson’s disease—the iron-related degenerative disease. Ageing Res Rev. 2024;101:102477. doi:10.1016/j.arr.2024.102477

9. Feng H, Schorpp K, Jin J, et al. Transferrin receptor is a specific ferroptosis marker. Cell Rep. 2020;30(10):3411–3423.e3417. doi:10.1016/j.celrep.2020.02.049

10. Ding K, Liu C, Li L, et al. Acyl-CoA synthase ACSL4: an essential target in ferroptosis and fatty acid metabolism. Chin Med J. 2023;136(21):2521–2537. doi:10.1097/CM9.0000000000002533

11. Tang F, Zhou LY, Li P, et al. Inhibition of ACSL4 alleviates parkinsonism phenotypes by reduction of lipid reactive oxygen species. Neurotherapeutics. 2023;20(4):1154–1166. doi:10.1007/s13311-023-01382-4

12. Chen X, Li J, Kang R, Klionsky DJ, Tang D. Ferroptosis: machinery and regulation. Autophagy. 2021;17(9):2054–2081. doi:10.1080/15548627.2020.1810918

13. Costa I, Barbosa DJ, Benfeito S, et al. Molecular mechanisms of ferroptosis and their involvement in brain diseases. Pharmacol Ther. 2023;244:108373. doi:10.1016/j.pharmthera.2023.108373

14. Song LM, Xiao ZX, Zhang N, et al. Apoferritin improves motor deficits in MPTP-treated mice by regulating brain iron metabolism and ferroptosis. iScience. 2021;24(5):102431. doi:10.1016/j.isci.2021.102431

15. Liu Y, Su Z, Tavana O, Gu W. Understanding the complexity of p53 in a new era of tumor suppression. Cancer Cell. 2024;42(6):946–967. doi:10.1016/j.ccell.2024.04.009

16. Song B, Yang P, Zhang S. Cell fate regulation governed by p53: friends or reversible foes in cancer therapy. Cancer Commun. 2024;44(3):297–360.

17. Liu Y, Gu W. p53 in ferroptosis regulation: the new weapon for the old guardian. Cell Death Differ. 2022;29(5):895–910. doi:10.1038/s41418-022-00943-y

18. Pan J, Xiong W, Zhang A, et al. The imbalance of p53-Park7 signaling axis induces iron homeostasis dysfunction in doxorubicin-challenged cardiomyocytes. Adv Sci. 2023;10(15):e2206007. doi:10.1002/advs.202206007

19. Li S, Wang M, Wang Y, et al. p53-mediated ferroptosis is required for 1-methyl-4-phenylpyridinium-induced senescence of PC12 cells. Toxicol In Vitro. 2021;73:105146. doi:10.1016/j.tiv.2021.105146

20. Yao Z, Jia F, Wang S, et al. The involvement of IRP2-induced ferroptosis through the p53-SLC7A11-ALOX12 pathway in Parkinson’s disease. Free Radic Biol Med. 2024;222:386–396. doi:10.1016/j.freeradbiomed.2024.06.020

21. Hu Q, Yao J, Wu X, et al. Emodin attenuates severe acute pancreatitis-associated acute lung injury by suppressing pancreatic exosome-mediated alveolar macrophage activation. Acta Pharm Sin B. 2022;12(10):3986–4003. doi:10.1016/j.apsb.2021.10.008

22. Mitra S, Anjum J, Muni M, et al. Exploring the journey of emodin as a potential neuroprotective agent: novel therapeutic insights with molecular mechanism of action. Biomed Pharmacother. 2022;149:112877. doi:10.1016/j.biopha.2022.112877

23. Park S, Park CW, Eom JH, et al. Preclinical and dose-ranging assessment of hESC-derived dopaminergic progenitors for a clinical trial on Parkinson’s disease. Cell Stem Cell. 2024;31(1):25–38.e28. doi:10.1016/j.stem.2023.11.009

24. Lei L, Yuan J, Dai Z, et al. Targeting the labile iron pool with engineered DFO nanosheets to inhibit ferroptosis for Parkinson’s disease therapy. Adv Mater. 2024;36(41):e2409329. doi:10.1002/adma.202409329

25. Hu S, Zhou J, Hao J, et al. Emodin ameliorates doxorubicin-induced cardiotoxicity by inhibiting ferroptosis through the remodeling of gut microbiota composition. Am J Physiol Cell Physiol. 2024;326(1):C161–c176. doi:10.1152/ajpcell.00477.2023

26. Jackson-Lewis V, Przedborski S. Protocol for the MPTP mouse model of Parkinson’s disease. Nat Protoc. 2007;2(1):141–151. doi:10.1038/nprot.2006.342

27. Liu H, Wu H, Zhu N, et al. Lactoferrin protects against iron dysregulation, oxidative stress, and apoptosis in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced Parkinson’s disease in mice. J Neurochem. 2020;152(3):397–415. doi:10.1111/jnc.14857

28. Kim TW, Koo SY, Riessland M, et al. TNF-NF-κB-p53 axis restricts in vivo survival of hPSC-derived dopamine neurons. Cell. 2024;187(14):3671–3689.e3623. doi:10.1016/j.cell.2024.05.030

29. Tang LJ, Zhou YJ, Xiong XM, et al. Ubiquitin-specific protease 7 promotes ferroptosis via activation of the p53/TfR1 pathway in the rat hearts after ischemia/reperfusion. Free Radic Biol Med. 2021;162:339–352. doi:10.1016/j.freeradbiomed.2020.10.307

30. Chen C, Huang Y, Xia P, et al. Long noncoding RNA Meg3 mediates ferroptosis induced by oxygen and glucose deprivation combined with hyperglycemia in rat brain microvascular endothelial cells, through modulating the p53/GPX4 axis. Eur J Histochem. 2021;65(3). doi:10.4081/ejh.2021.3224

31. Guo Y, Chen Y, Zhang H, et al. Emodin attenuates hypoxic-ischemic brain damage by inhibiting neuronal apoptosis in neonatal mice. Neuroscience. 2024;554:83–95. doi:10.1016/j.neuroscience.2024.06.030

Specialty

Please chooseI’m not a medical professional.Allergy and ImmunologyAnatomyAnesthesiologyBiostatisticsCardiac/Thoracic/Vascular SurgeryCardiologyCritical CareDentistryDermatologyDiabetes and EndocrinologyEmergency MedicineEpidemiology and Public HealthFamily MedicineForensic MedicineGastroenterologyGeneral PracticeGeneticsGeriatricsHealth PolicyHematologyHIV/AIDSHospital-based MedicineI’m not a medical professional.Infectious DiseaseIntegrative/Complementary MedicineInternal MedicineInternal Medicine-PediatricsMedical Education and SimulationMedical PhysicsMedical StudentNephrologyNeurological SurgeryNeurologyNuclear MedicineNutritionObstetrics and GynecologyOccupational HealthOncologyOphthalmologyOptometryOral MedicineOrthopaedicsOsteopathic MedicineOtolaryngologyPain ManagementPalliative CarePathologyPediatricsPediatric SurgeryPharmacologyPhysical Medicine and RehabilitationPlastic SurgeryPodiatryPreventive MedicinePsychiatryPsychologyPulmonologyRadiation OncologyRadiologyRheumatologySubstance Use and AddictionSurgeryTherapeuticsTraumaUrologyMiscellaneous

HSBC and Mistral AI have announced a strategic partnership to enhance and accelerate the use of generative AI across the bank, improving business processes, saving employees time and helping to better serve millions of customers globally.

This multi-year partnership provides access to Mistral AI’s commercial models, including future developments.

Under the agreement, HSBC and Mistral’s applied AI, science and engineering teams will collaborate on the development of generative AI solutions across the organisation.

HSBC will combine its strong internal technology capabilities with Mistral AI’s deep expertise in foundational model development. This will enable HSBC to enhance current AI initiatives through self-hosted AI models that operate on HSBC’s internal technology systems.

HSBC is continually assessing a wide range of large language models (LLMs) as part of its technology strategy. The bank saw a valuable opportunity to use Mistral’s AI expertise to enhance its internal tools, including an AI-powered platform used by HSBC colleagues globally to help with productivity tasks. This includes:

• creating business tasks which support a variety of needs across the bank, such as allowing client-facing teams to deliver tailored communications at speed, enabling marketing teams to launch hyper-personalised campaigns, and helping procurement teams identify risks and savings opportunities
• enhancing financial analysis of complex and document-heavy client lending or financing processes
• multilingual reasoning and translation services: helping to translate and validate information in multiple languages to inform customer interactions
• faster development innovation cycles: enabling teams to prototype, validate and launch new processes or features more rapidly

The partnership with Mistral builds on HSBC’s investments in the latest AI technologies which are focused on increasing business efficiencies and better serving customers.

Future areas of focus for HSBC will include customer-facing innovations, such as improvements to credit and lending processes, enhancing customer onboarding, and fraud and anti-money laundering checks.

Georges Elhedery, Group CEO, HSBC, said:

“Working with Mistral is an exciting step forward in HSBC’s technology strategy, enabling us to further enhance AI capabilities across the bank. The partnership will equip our colleagues with tools to help them innovate, simplify daily tasks, and free up time to deliver for our customers.”

Both organisations are committed to the responsible use of AI and will work together to ensure all deployments adhere to the highest standards of AI transparency, data privacy, and technology development.

Arthur Mensch, Mistral AI Co-Founder and Chief Executive Officer, said:

“We are proud to engage in this long-term partnership with HSBC. Our highly customisable, enterprise-grade frontier AI solutions will reinvent HSBC’s workflows and services while ensuring full ownership of data. Together, we will provide HSBC’s employees with high-end, AI-powered productivity tools and a new generation of banking services for millions of customers worldwide.”