CMV-infection, plaques and high blood pressure

26 05 2009

A recent experimental study shows that a common virus that hides (becomes latent) after infection can trigger high blood pressure, a leading cause of cardiovascular disease.

The virus is cytomegalovirus (CMV). It belongs to the herpesviruses, hence its alternative name Human Herpesvirus 5 (HHV-5). The herpesviruses have in common that they can become latent (dormant) for a long while. Herpes simplex virus-1 (HSV-1) and varizella zoster virus usually hide in nerves (neurons) and can cause cold sores and shingles respectively after reactivation.

Cytomegalovirus (CMV) can be transmitted by sputum and other bodily fluids. After infection, it stays within the host, hiding primarily in certain white blood cells serving as a reservoir for reactivation.

The study by Cheng et al, consisted of an experimental mouse study and in vitro experiments, using mouse CMV (MCMV) and human CMV (HCMV)

  1. In the in vivo experiment, 4 groups of mice -of 12 animals each- were treated as follows:
    • V: Mice in group 1 were infected by MCMV and fed a regular diet.
    • M: Mice in group 2 were mock-infected and fed a regular diet (control group).
    • V-HD: Mice in group 3 were infected by MCMV and fed a high cholesterol diet.
    • HD: mock-infected and fed a high cholesterol diet (control group).
      systolic diastolic
  2. After 6 weeks the blood pressure, measured in the carotid artery, was significantly enhanced in mice that were infected by the MCMV compared to mock-infected controls. Thus the blood pressure was significantly higher in V versus M and in V-HD versus HD mice. This was both true for the systolic and the diastolic blood pressure (see Fig).
  3. Atherosclerotic plaque formation in mouse aortas was only observed in V-HD mice (MCMV infection and high cholesterol diet together). 3 out of 12 mice showed plaques against 0 out of 12 mice in the other groups.
  4. The serum levels of all 3 pro-inflammatory cytokines tested, IL-6, TNF-α and MCP-1, were higher in the mice infected with MCMV (V-HD and V) than in the mock-infected groups fed with either diet.
  5. CMV infection induced renin and Ang II expression (in vitro cq in vivo). Renin is a rate limiting protein of Renin-Ang-II system and Ang II is the peptide that causes the blood vessels to constrict, thereby enhancing blood pressure.
  6. The renin production was only enhanced in kidney-cells containing CMV-particles.
  7. In mice infected with CMV, Ang II was also increased in serum and in artery tissues.
  8. MCMV RNA and DNA were present in blood vessels post infection. Mice produced the latency-related CMV transcriptional product I-E1.
  9. Next, in vitro experiments were carried out with human cells. Infection of human blood vessel endothelial cells with HCMV induced renin expression in a-dose dependent manner, without lysis of the cells (non-lytic infection). This means that viral gene expression takes place in the absence of a viral cytopathic effect. Only replicating forms of HCMV could induce renin.

Conclusion of the Authors:

“In summary, CMV infection alone caused a significant increase of arterial blood pressure. Enhanced expression of pro-inflammatory cytokines, renin and Ang II underlies the pathogenic mechanism of an active CMV infection to increase blood pressure and aggravate atherosclerosis. Thus, control of CMV infection to restrict development of hypertension and atherosclerosis may provide a new strategy to prevent cardiovascular diseases associated with HCMV infection.”

The present article provides interesting new insights into various mechanisms whereby CMV could ultimately cause cardiovascular disease. Many different assays were performed, both with mouse CMV in vivo and in vitro and with human CMV in vitro.

It should be noted however, that several of the reported findings are not new. CVM DNA and RNA have long been identified in atherosclerotic plaques. CMV and other herpesviruses were known to induce IL-6 and other pro-inflammatory interleukines. And a relation between CMV-infection and cardiovascular disease has been established in the past.

The main new findings are (1) that CVM-infection can raise blood pressure and (2) the demonstration of possible mechanisms involved in the increase of blood pressure: induction of renin and angiotensin by infection of endothelial and renal cells (rather than induction of plaques?) and induction of cytokines. Importantly, the cytokine-enhancing effects were restricted to CMV-infected cells only.

Although the findings are very suggestive of a causal relationship, it remains to be proven whether hypertension via enhanced extrarenal production of renin and angiotensin is the causal factor of CMV-associated cardiovascular disease in the normal human situation. Some statements are not underpinned by the current data. See for instance the blue marked text in the author’s conclusion.


Arterial wall depicting major inflammatory features common to CMV infection and atherosclerosis. – K. Froberg

The current prevailing hypothesis is that atherogenesis is an inflammatory response to acute or chronic endothelial injury. Possible causes for this endothelial injury include free radicals, modified LDL, hypertension, diabetes mellitus and smoking [4,5]. More recently CMV and other infections have been added to the list of risk factors. It is generally thought that the formation of plaques concurrent with the enhanced induction of cytokines can contribute to atherosclerosis and vascular disease, after many years of chronic, active inflammation of the arterial wall.

In the present model, atherosclerosis was only observed in 3 out of 12 HD-V mice. This indicates that atherosclerosis is not involved in the increase of blood pressure. Instead, the present study shows, that CMV might contribute to cardiovascular disease via an effect on blood pressure, presumably mediated through an enhanced expression of renin and Ang II.

With regard to the statistical analysis, I wonder why each virus-treated group was only compared to its mock-infected control (using the Student’s t-test)? Why weren’t all groups compared to each other, using another statistical test? Is this because another approach would have rendered most differences insignificant? The differences between groups are consistent but not very large and the standard deviations are quite overlapping. (Thus significance might be lost when more rigid test are used)

It would for instance be interesting to know the relative contribution of the fatty diet to the rise in blood pressure, or the risk in the 4 groups relative to each other.
We all have CMV (well about 70-90%), but just few of us get high blood pressure (at young age). So likely other factors come into play, that may enhance the detrimental effects of CMV or vice versa. Therefore, it would for instance be interesting to know, to which extent high fat consumption would contribute to high blood pressure as well. Is the suggested order as we see it real: V alone giving slight elevation of blood pressure, HD alone giving more elevation and VHD giving the highest rise? (2×2 table).

The study has some fascinating results, but it is too soon to think about antiviral treatment as the authors and media [3] suggest. Nor is it necessary to worry the public by stating that: You can find out if you have been infected with CMV by requesting a blood test from your GP”, as the daily mail does. This is useless, because CMV is omnipresent.

It hardly seems to be justified to suggest that everybody should be treated with antivirals. And what about the other viruses or bacteria that may play a role? Furthermore, lifestyle and other (immunological) parameters may be more important ànd lifestyle changes may be easier to implement.

Update: Tuesday 2009-05-26: 17.30 pm.

HATTIP: @microbytes (Twitter)

Figure: Kent Froberg from the University of Minesota (


  1. Cheng, J., Ke, Q., Jin, Z., Wang, H., Kocher, O., Morgan, J., Zhang, J., & Crumpacker, C. (2009). Cytomegalovirus Infection Causes an Increase of Arterial Blood Pressure PLoS Pathogens, 5 (5) DOI: 10.1371/journal.ppat.1000427
  4. Many reviews from Cathrien A. Bruggeman and colleagues, like this review in a Book.
  5. Kent Froberg from the University of Minesota (
Human Herpesvirus 5 (HHV-5)



6 responses

26 05 2009
Walter Jessen

A t-test was used to compare each virus-treated group to its mock-infected control because an ANOVA against all four groups wouldn’t identify group-to-group differences.

Your issue with the error bars is valid if they were SEMs. Unfortunately, they’re SDs. SDs don’t account for sample size, so whether or not the error bars overlap tells you nothing about the difference.

I’m not sure what you mean about V-HD having a greater effect than V. In terms of overall blood pressure, it’s higher because the baseline was higher (high cholesterol diet). The difference between HD and V-HD may be slightly higher than the difference between M and V, but it’s likely due to the larger SD for V-HD than for V.

Thanks, interesting study.

26 05 2009
Grand Rounds « See First Blog

[…] Laika’s MedLib Blog, Jacqueline posts about research showing a certain virus might cause high blood […]

26 01 2010

It seems interesting that a virus that uses arginine to proliferate aslo maybe involved with cardiovascular disease. It would be interesting to find out if it is robbing the body of arginine there by reducing the nitric oxide needed for healthy endothelium cells.

6 04 2010
cardio cocktail

Since so much of the population has CMV what makes the difference for some to get hypertension and other to not? Could the amino acid lysine be a supplement that reduces the effect of the CMV virus since it can reduce the out break of herpes?

10 01 2013
Edwin Rahr

Crucial Error – In section 1 you’ve said that M group was infected with MCMV and that V group was mock infected. This is not the case!
From the original study ” V: mice infected by MCMV and fed with a regular diet. M: mice mock-infected and fed with a regular diet.”

This is a very important difference!

11 01 2013

You are right. M and V are mixed up in the legends/introduction) (above the figure), but not in the text. Although this is a clear mistake, it has no effect on the findings and the conclusions (abbreviations are properly explained in the remainder of the text).

Thanks I will replace M by V and vice versa.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: